68 views

Uploaded by kashif salman

save

- Handout Eight
- Stability Assessment of a Critical Rock Slope
- CFA Exam Result Explanation
- Risk Reliability Summary and Tables
- Efficiency of the Philippine Stock Market
- 39 3 Sums n Diffs Rndm Vars
- Statistics for business 5
- Condicion de Mantto y Descripcion Del Sistema
- Introduction
- Markov Chain Exercises
- Statistics and Data With R
- Chapter_11 Sampling Distributions and Estimation
- Thoughts About Estimation
- 3-2Grading
- Pert
- Wikipedia - Standar Deviasi
- 601Example Statistic Project
- Chapter 8 and 9 Stats Review
- MKT RES 5-6
- Formula Sheet Non-cumulative Final F2011
- Handout7
- CH17 Mock Test
- Probability, Statistics and Stochastic Processes
- MIL-STD-781C and confidence intervals on MTBF.pdf
- a069966.pdf
- 5 Biomechanics
- statsproject
- Introduction to Statistics With SAS
- Week8Assessment
- Nasa/Tm—2005 213958
- f.docx
- ucsf
- Glosario de Terminos Microeconomicos
- Dibujos_Fuera_del_Papel.pdf
- PLANTAS MEDICINAIS - Coletânia de saberes.pdf
- Lab # 2A Introduction to Induction Motor.
- Puntos Control
- 3336-1574-1-PB
- WORKSHEET IPA
- penelitian herbisida
- KUESIONER PENELITIAN siap
- INNOVAplegV5.pdf
- Cobertura Curric. Cs. Naturales
- El Método Seger en Excel- Instrucciones de Uso. Guía de Formulación Cerámica (Seminario de Investigación)
- cv r1
- r robertson.pdf
- antebellum periodicals timeline
- 05.04.2014 a Sociedade Tolera Agressão Sexual Às Mulheres
- BAB II ADL
- Testigos de Jehova
- 35007b-75235
- df
- combo1mata.pdf
- trowardt2563825638-8.pdf
- grcevic_kopitar
- Cofer
- Carl Gustav Jung - Amintiri, vise, reflectii.pdf
- 0p.docx
- Upload
- 360129114-PROGRAM-KERJA-WAKIL-KEPALA-SEKOLAH-URUSAN-KESISWAAN-doc.doc
- leadership preferences in japan 2007
- organisational knowledge leadership- grounded theory approach-2007
- sunflower
- challenges and prospects of hrm practices in developing countries
- organisational support, employee development and commitment
- leadership theory
- gender and ownership in uk small firms
- leadership of a culture change process
- Introduction to Microsoft Office
- women enterpreneurs- management skills and business problems
- Children Attitude Toward TVC in Pak
- imapct of selected hr parctices on percieved emplyee performance
- transformational leadership
- organisational commitment among pakistani university teachers
- uk university article
- article12
- markus
- appraisal2
- holy places
- article4
- afshan 4
- max planck
- advertising bans
- appraisal
- advertising under varticle producat differenate
- appraisal employee development
- appraisal1
- advertising create a goodwill of producate
- advertising planning
- alien definition

You are on page 1of 23

**163–185, October 2004 (© 2004)
**

Competitive Advertising under Uncertainty:

A Stochastic Differential Game Approach

1

A. Prasad

2

and S. P. Sethi

3

Communicated by G. Leitmann

Abstract. We analyze optimal advertising spending in a duopolistic

market where each ﬁrm’s market share depends on its own and its

competitor’s advertising decisions, and is also subject to stochastic

disturbances. We develop a differential game model of advertising in

which the dynamic behavior is based on the Sethi stochastic advertis-

ing model and the Lanchester model of combat. Particularly impor-

tant to note is the morphing of the sales decay term in the Sethi

model into decay caused by competitive advertising and noncompeti-

tive churn that acts to equalize market shares in the absence of adver-

tising. We derive closed-loop Nash equilibria for symmetric as well as

asymmetric competitors. For all cases, explicit solutions and compar-

ative statics are presented.

Key Words. Advertising, dynamic duopoly, competitive strategy,

differential games, stochastic differential equations.

1. Introduction

The advertising spending decision has been the focus of considerable

interest for researchers in marketing as evidenced by the large body of lit-

erature devoted to this subject starting with Vidale and Wolfe (Ref. 1) and

reported in surveys by Sethi (Ref. 2) and by Feichtinger, Hartl, and Sethi

(Ref. 3). The annual expenditure on advertising by ﬁrms was 117 billion

dollars in 2002 in the US alone (Ref. 4). At the same time, marketers have

1

The authors thank the PhD seminar participants at the University of Texas at Dallas

for helpful comments.

2

Assistant Professor, School of Management, University of Texas at Dallas, Richardson,

Texas.

3

Ashbel Smith Professor, School of Management, University of Texas at Dallas,

Richardson, Texas.

163

0022-3239/04/1000-0163/0 © 2004 Plenum Publishing Corporation

164 JOTA: VOL. 123, NO. 1, OCTOBER 2004

noted that ﬁrms advertise often in a suboptimal manner. For example,

Patti and Blasko (Refs. 5–6) found in surveys that a large percentage of

industrial and consumer good ﬁrms do advertising budgeting based on the

affordable method, the percentage-of-sales method, and the competitive-

parity method. These methods are rarely optimal. A number of researchers

have concluded also that ﬁrms tend to overadvertise (Refs. 7–8).

The combination of large amounts of money spent on advertising and

potential inefﬁciencies in the advertising budgeting process motivates the

need to better understanding optimal advertising budgeting. However, one

must be careful to limit the conclusions of optimality only to those mar-

kets for which the model applies. Thus, we deﬁne our market context and

research question as follows.

We examine a duopoly market in a mature product category where

two ﬁrms compete for market share using advertising as the dominant

marketing tool. The ﬁrms are strategic in their behavior; that is, they take

actions that maximize their objective while considering also the actions

of the competitor. Furthermore, they interact dynamically for the fore-

seeable future. This is due in part to the carryover effect of advertising,

which means that advertising spending today will continue to inﬂuence

sales several months down the line. Each ﬁrm’s advertising acts to increase

its market share, while the competitor’s advertising acts to reduce its mar-

ket share. In addition to market share decay caused by competitive adver-

tising, noncompetitive factors described in the next paragraph can cause

market share churn. However, marketing and competitive activities alone

do not govern market shares in a deterministic manner, because there is

inherent randomness in the marketplace and in the choice behavior of cus-

tomers. The market for cola drinks, dominated by Coke and Pepsi, is an

example of a market with such features (Refs. 9–11).

Explanations for churn are product obsolescence, forgetting (Ref. 1),

lack of market differentiation (Ref. 12), lack of information (Ref. 13), vari-

ety seeking (Ref. 14), and brand switching. These factors do not neces-

sarily cause market share to decay because the decay of market share for

one ﬁrm is a gain in market share for the other. Hence, we use the term

“churn” rather than “decay”. Due to churn, the market shares converge to

a long-run equilibrium when neither brand is advertised for a very long

duration. The proposed model takes into account decay due to competi-

tive advertising as well as churn due to noncompetitive factors.

For a competitive market with stochastic disturbances and other fea-

tures as described above, our objective is to recommend optimal advertis-

ing expenditures over time for the two ﬁrms. A stochastic differential game

model is formulated based on the monopoly model of Sethi (Ref. 15). We

obtain explicit closed-loop solutions.

JOTA: VOL. 123, NO. 1, OCTOBER 2004 165

Our research follows in the operations research tradition in marketing

(Refs. 9–11 and Refs. 16–19). Whereas elements of the marketing envi-

ronment described above, such as dynamics, competition, competitive and

noncompetitive decay, and also stochasticity, have been commonly used

by individual models, there exist few attempts to study them together. We

provide a focused discussion of the literature in Section 2.

2. Background

Among the earliest aggregate response models is the Vidale-Wolfe

model whose dynamics is given by

dx(t )/dt =ρu(t )(1−x(t )) −δx(t ), x(0) =x

0

, (1)

where x(t ) is the sales rate (expressed as a fraction of the total market)

at time t, u(t ) is the advertising expenditure rate, ρ is a response constant,

and δ is a market share decay constant; ρ determines the effectiveness of

advertising, while δ determines the rate at which consumers are lost due to

product obsolescence, forgetting, etc. The formulation has several desirable

properties; for example, market share has a concave response to advertis-

ing, and there is a saturation level (Ref. 20). The optimal advertising path

for the Vidale-Wolfe dynamics is provided by Sethi (Ref. 21).

Subsequent research extended the basic framework to incorporate

competitive advertising (Refs. 22–23). A competitive extension of the

Vidale-Wolfe model, based on the Lanchester model of combat (Refs. 20

and 24), is

dx(t )/dt =ρ

1

u

1

(x(t ), y(t ))(1−x(t ))

−ρ

2

u

2

(x(t ), y(t ))x(t ), x(0) =x

0

, (2)

dy(t )/dt =ρ

2

u

2

(x(t ), y(t ))x(t )

−ρ

1

u

1

(x(t ), y(t ))(1−x(t )), y(0) =1−x

0

, (3)

where x(t ) and y(t ) represent the market shares of the two ﬁrms, whose

parameters and decision variables are indexed 1 and 2 respectively. Note

that

x(t ) +y(t ) =1.

In addition to competitive extensions, recent research has examined

the problem where the state variable, usually market share, is determined

by stochastic disturbances in addition to advertising spending. We start

166 JOTA: VOL. 123, NO. 1, OCTOBER 2004

with the stochastic, monopoly advertising model of Sethi (Ref. 15). The

Sethi model is given by the It ˆ o equation

dx(t ) =

ρu(t )

1−x(t ) −δx(t )

dt +σ(x)dw(t ), x(0) =x

0

, (4)

where σ(x) represents a variance term and w(t ) represents a standard

Wiener process. The model has the useful feature in that it has a basic

resemblance to the Vidale-Wolfe model and at the same time it permits an

explicit solution to the advertising spending decision. We wish to extend

this model to incorporate competition.

A related extension is due to Sorger (Ref. 19). He uses a special case

of the Lanchester model to obtain a duopoly version of the Sethi model.

This is

dx/dt =ρ

1

u

1

(x, y)

√

1−x −ρ

2

u

2

(x, y)

√

x, x(0) =x

0

, (5)

dy/dt =ρ

2

u

2

(x, y)

1−y −ρ

1

u

1

(x, y)

√

y, y(0) =1−x

0

. (6)

Chintagunta and Jain (Ref. 17) test this speciﬁcation using data from

the pharmaceuticals, soft drinks, beer and detergent industries, and ﬁnd

it to be appropriate. Sorger describes the appealing characteristics of the

speciﬁcation, noting that it is compatible with word-of-mouth and nonlin-

ear effects. However, the decay constant δ is not included here and it is

assumed to be replaced totally by competitive effects.

On the other hand, we extend the Sethi model to allow for competi-

tion. We are able to do so, while retaining the decay constant. Note that a

stochastic version of the Sorger model is a special case of ours (speciﬁed

in the next section) when δ =0. The decay constant, which goes back to

the Vidale-Wolfe formulation, is not solely replaced by competitive adver-

tising effects. Thus, in order to capture effects such as forgetting, the decay

parameter has been morphed into the churn parameter. We will discuss

how including the term affects the outcome.

3. Model

We consider a duopoly market in a mature product category where

total sales are distributed between two ﬁrms, denoted ﬁrm 1 and ﬁrm 2,

which compete for market share through advertising spending. We denote

the market shares of ﬁrms 1 and 2 at time t as x(t ) and y(t ), respectively.

Using the subscript i ∈{1, 2} to reference the two ﬁrms, u

i

(x(t ), y(t ), t ) ≥0

is the advertising control, ρ

i

>0 is the advertising effectiveness parameter,

r

i

>0 is the discount rate, δ >0 is the churn parameter, and c

i

>0 is a cost

JOTA: VOL. 123, NO. 1, OCTOBER 2004 167

parameter. The time argument will be suppressed in future where no con-

fusion arises. The model dynamics is given by

dx =[ρ

1

u

1

(x, y)

√

1−x −ρ

2

u

2

(x, y)

√

x −δ(x −y)]dt

+σ(x, y)dw, x(0) =x

0

, (7)

dy =[ρ

2

u

2

(x, y)

1−y −ρ

1

u

1

(x, y)

√

y −δ(y −x)]dt

−σ(x, y)dw, y(0) =1−x

0

. (8)

This speciﬁcation has the same desirable properties of concave response

with saturation as the Vidale-Wolfe model. The market share is nonde-

creasing with own advertising and nonincreasing with the competitor’s

advertising expenditure. Consistent with the literature, churn is propor-

tional to the market share. As discussed, it is caused by inﬂuences other

than competitive advertising, such as a lack of perceived differentiation

between brands, so that the market shares tend to converge in the absence

of advertising. Finally, the market shares are subject to a white noise term

σ(x, y)dw.

Since

dx +dy =0 and x(0) +y(0) =1,

we have

x(t ) +y(t ) =1, for all t ≥0.

Now that y(t ) = 1 − x(t ), we need use only the market share of ﬁrm

1 to completely describe the market dynamics. Thus, u

i

(x, y), i = 1, 2,

and σ(x, y) can be written as u

i

(x, 1 −x) and σ(x, 1 −x). With a slight

abuse of notation, we will use u

i

(x) and σ(x) in place of u

i

(x, 1−x) and

σ(x, 1−x), respectively. Thus,

dx =[ρ

1

u

1

(x)

√

1−x −ρ

2

u

2

(x)

√

x −δ(2x −1)]dt +σ(x)dw,

x(0) =x

0

, (9)

with 0≤x

0

≤1.

As noted in Ref. 15, when choosing a formulation, an important con-

sideration is that the market share should remain bounded within [0, 1],

which can be problematic given stochastic disturbances. In our model, it is

easy to see that x ∈[0, 1] almost surely for t >0, as long as u

i

(x) and σ(x)

are continuous functions that satisfy the Lipschitz conditions on every

closed subinterval of (0, 1) and further that u

i

(x) ≥0, x ∈[0, 1], and

σ(x) >0, x ∈(0, 1) and σ(0) =σ(1) =0. (10)

168 JOTA: VOL. 123, NO. 1, OCTOBER 2004

With this, we have a strictly positive drift at x =0 and a strictly negative

drift at x =1, i.e.,

ρ

1

u

1

(0)

√

1−0+δ >0 and −ρ

2

u

2

(1) −δ <0. (11)

Then, from Gihman and Skorohod (Ref. 25, Theorem 2, pp. 149 and

157–158), x =0 and x =1 are natural boundaries for the solutions of (9)

with x

0

∈[0, 1]; i.e., x ∈(0, 1) almost surely for t >0.

Let m

i

denote the industry sales volume multiplied by the per-unit

proﬁt margin for ﬁrm i. We formulate the optimal control problem faced

by the two ﬁrms as

Max

u

1

≥0

¸

V

1

(x

0

) =E

∞

0

e

−r

1

t

[m

1

x(t ) −c

1

u

1

(t )

2

]dt

¸

, (12)

Max

u

2

≥0

¸

V

2

(x

0

) =E

∞

0

e

−r

2

t

[m

2

(1−x(t )) −c

2

u

2

(t )

2

]dt

¸

, (13)

s.t., dx =[ρ

1

u

1

(x)

√

1−x −ρ

2

u

2

(x)

√

x −δ(2x −1)]dt +σ(x)dw, (14)

x(0) =x

0

∈[0, 1]. (15)

Each ﬁrm seeks to maximize its expected, discounted proﬁt stream subject

to the market share dynamics. Note that, when the advertising expenditure

enters linearly in the dynamic equation, its cost in the objective function

is assumed to be quadratic as in Ref. 15. Equivalently, one can take the

square root of the advertising expenditure in the dynamic equation and

subtract the advertising expenditure linearly in the objective function (e.g.,

Ref. 19). See Refs. 26–27 for a discussion.

4. Analysis

To ﬁnd the closed-loop Nash equilibrium strategies, we form the

Hamilton-Jacobi-Bellman (HJB) equation for each ﬁrm,

r

1

V

1

=max

u

1

¸

m

1

x −c

1

u

2

1

+V

1

(ρ

1

u

1

√

1−x

−ρ

2

u

∗

2

√

x −δ(2x −1)) +σ(x)

2

V

1

/2

¸

, (16)

r

2

V

2

=max

u

2

¸

m

2

(1−x) −c

2

u

2

2

+V

2

(ρ

1

u

∗

1

√

1−x

−ρ

2

u

2

√

x −δ(2x −1)) +σ(x)

2

V

2

/2

¸

, (17)

where V

i

=dV

i

/dx, V

i

=d

2

V

i

/dx

2

and where u

∗

1

and u

∗

2

denote the com-

petitor’s advertising policies in (16) and (17), respectively. We obtain the

JOTA: VOL. 123, NO. 1, OCTOBER 2004 169

optimal feedback advertising decisions

u

∗

1

(x) =max

0, V

1

(x)ρ

1

√

1−x/2c

1

, (18a)

u

∗

2

(x) =max

0, −V

2

(x)ρ

2

√

x/2c

2

. (18b)

Since 0 ≤x ≤1 and since it is reasonable to expect V

1

≥0 and V

2

≤0, we

can reduce the advertising decisions (18) to

u

∗

1

(x) =V

1

(x)ρ

1

√

1−x/2c

1

, (19a)

u

∗

2

(x) =−V

2

(x)ρ

2

√

x/2c

2

, (19b)

which hold as we shall see later. Substituting (19) in (16) and (17), we

obtain the Hamilton-Jacobi equations

r

1

V

1

=m

1

x +V

2

1

ρ

2

1

(1−x)/4c

1

+V

1

V

2

ρ

2

2

x/2c

2

−V

1

δ(2x −1) +σ(x)

2

V

1

/2, (20)

r

2

V

2

=m

2

(1−x) +V

2

2

ρ

2

2

x/4c

2

+V

1

V

2

ρ

2

1

(1−x)/2c

1

−V

2

δ(2x −1) +σ(x)

2

V

2

/2. (21)

Following Ref. 15, we obtain the following forms for the value functions:

V

1

=α

1

+β

1

x, (22a)

V

2

=α

2

+β

2

(1−x). (22b)

These are inserted into (20) and (21) to determine the unknown coefﬁ-

cients α

1

, β

1

, α

2

, β

2

. Equating powers of x in Eq. (20) and powers of 1−x

in Eq. (21), the following four equations emerge, which can be solved for

the unknown coefﬁcients:

r

1

α

1

=β

2

1

ρ

2

1

/4c

1

+β

1

δ, (23)

r

1

β

1

=m

1

−β

2

1

ρ

2

1

/4c

1

−β

1

β

2

ρ

2

2

/2c

2

−2β

1

δ, (24)

r

2

α

2

=β

2

2

ρ

2

2

/4c

2

+β

2

δ, (25)

r

2

β

2

=m

2

−β

2

2

ρ

2

2

/4c

2

−β

1

β

2

ρ

2

1

/2c

1

−2β

2

δ. (26)

Since for ﬁrms having different parameter values, the solutions to

these equations are complicated, we consider ﬁrst the case of symmetric

ﬁrms in Section 4.1. The case of asymmetric ﬁrms will be dealt with in

Section 4.2.

170 JOTA: VOL. 123, NO. 1, OCTOBER 2004

4.1. Symmetric Firms. For the symmetric ﬁrm case,

α =α

1

=α

2

,

β =β

1

=β

2

,

m=m

1

=m

2

,

c =c

1

=c

2

,

ρ =ρ

1

=ρ

2

,

r =r

1

=r

2

.

The four equations in (23)–(26) reduce to the following two:

rα =β

2

ρ

2

/4c +βδ, (27a)

rβ =m−3β

2

ρ

2

/4c −2βδ. (27b)

There are two solutions for β. One is negative, which clearly makes no

sense. Thus, the remaining positive solution is the correct one. This gives

also the corresponding α. The solution is

α =

(r −δ)(W −

W

2

+12Rm) +6Rm

/18Rr, (28a)

β =

W

2

+12Rm−W

/6R, (28b)

where

R=ρ

2

/4c, W =r +2δ.

We can see now that, with the solution for the value function, the con-

trols speciﬁed in Eq. (18) reduce to (19). This validates our choice of (19)

in deriving the value function.

Table 1 provides the comparative static results for the parameters

(proofs with the authors).

When ρ increases or c decreases (i.e., there is a marginal increase in

the value of advertising or a reduction in its cost), then as one might

expect, the amount of advertising increases. However, contrary to what

one would expect to see in a monopoly model of advertising, the value

function decreases. This occurs because, in this market, all advertising

occurs from competitive motivations, since the optimal advertising expen-

diture would be zero if a single ﬁrm were to own both identical products.

Advertising does not increase the size of the marketing pie, but affects

only its allocation. Thus, the increase in advertising causes a decrease

in the value function. The same logic does not apply when m increases

JOTA: VOL. 123, NO. 1, OCTOBER 2004 171

Table 1. Comparative statics for symmetric ﬁrms.

Variables Parameters

c ρ m δ r

α + − + + −

β + − + − −

u

∗

− + + − −

V(x) + − + ? −

Legend: increase (+), decrease (−), ambiguous (?).

or r decreases. In these cases, it is true that the wasteful advertising is

increased, but it is also true that the size of the pie is increased.

The churn parameter δ reduces competitive intensity. Hence, it might

be expected that an increase in δ should increase the proﬁtability by reduc-

ing advertising. In fact, only the constant α part of the value functions

increases, and it is ambiguous what happens to the value functions overall.

We can derive the exact conditions under which there is an increase or a

decrease in the value function of a ﬁrm due to an increase in δ. We ﬁnd

that, if the market share of a ﬁrm is less than half, the effect on the ﬁrm’s

value function is always positive. However, if the market share of a ﬁrm is

greater than half, its value function can decrease, because of an increase in

δ, if the following inequality is satisﬁed:

x >

¸

(r +2δ)

2

+12Rm−(r +2δ)

¸

6r +1/2.

The reason is that, when a ﬁrm has a market share advantage over its

rival, δ helps the rival unequally by tending to equalize market shares.

4.2. Asymmetric Firms. We return now to the general case of asym-

metric ﬁrms. For asymmetric ﬁrms, we reexpress equations (23)–(26) in

terms of a single variable β

1

, which is determined by the solution to the

equations

3R

2

1

β

4

1

+2R

1

(W

1

+W

2

)β

3

1

+

4R

2

m

2

−2R

1

m

1

−W

2

1

+2W

1

W

2

β

2

1

+2m

1

(W

1

−W

2

)β

1

−m

2

1

=0, (29)

α

1

=β

1

(β

1

R

1

+δ)/r

1

, (30)

β

2

=(m

1

−β

2

1

R

1

−β

1

W

1

)/2β

1

R

2

, (31)

α

2

=β

2

(β

2

R

2

+δ)/r

2

, (32)

172 JOTA: VOL. 123, NO. 1, OCTOBER 2004

where

R

1

=ρ

2

1

/4c

1

, R

2

=ρ

2

2

/4c

2

, W

1

=r

1

+2δ, W

2

=r

2

+2δ.

Once we obtain the correct value of β

1

out of the four solutions of the

quartic equation (29), the other coefﬁcients can be obtained by solving for

α

1

, β

2

, and then α

2

. The solution is given in Appendix A (Section 6).

We collect the main results of the analysis into Theorem 4.1.

Theorem 4.1. For the advertising game described by Eqs. (12)–(15):

(i) There exists a unique closed-loop Nash equilibrium solution to

the differential game. See proof in Appendix A, Section 6.

(ii) Optimal advertising is u

∗

1

(x) = β

1

ρ

1

√

1−x/2c

1

, u

∗

2

(x) =

β

2

ρ

2

1−y/2c

2

; in the symmetric ﬁrm case, β

1

= β

2

=

(

W

2

+12Rm − W)/6R; in the asymmetric ﬁrm case, β

1

and β

2

are given by (61) and (31).

We observe that the optimal advertising policy is to spend in propor-

tion to the competitor’s market share. Consistent with Ref. 19, the ﬁrm

that is in a disadvantageous position ﬁghts harder than its opponent and

should succeed in wrestling market share from the opponent. Spending

is decreasing in the own market share; thus, the advertising-to-sales ratio

is higher for the lower market share ﬁrm. As noted in the introduction,

many ﬁrms do advertising budgeting based on the affordable method, the

percentage-of-sales method, and the competitive-parity method (Refs. 5–6).

These methods suggest that the ﬁrm with lower market share should spend

less on advertising in contrast to the optimal advertising policy derived

here.

Table 2 provides the comparative static results for α, β, and V

1

(x)

with respect to the parameters, see proofs in Appendix B, Section 7.

Table 2. Comparative statics for asymmetric ﬁrms.

Variables Parameters

c

i

, c

j

ρ

i

, ρ

j

m

i

, m

j

δ r

i

, r

j

α

i

?, + ?, − +, − ? −, +

β

i

?, + ?, − +, − − −, +

u

∗

i

−, + +, − +, − − −, +

V

i

(x) ?, + ?, − +, − ? −, +

Legend: increase (+), decrease (−), ambiguous (?).

JOTA: VOL. 123, NO. 1, OCTOBER 2004 173

A comparison of Tables 1 and 2 shows the following features. First,

due to the complexity of the asymmetric case, more effects are ambigu-

ous. Second, a change in the own parameters has the same effect in the

asymmetric case as a change in these parameters for the symmetric case.

This is to be expected, since the ﬁrst-order effects likely dominate the sec-

ond-order effects, thus yielding the same results as in the symmetric case.

Third, a beneﬁcial change in the own parameters (ρ

i

, c

i

, m

i

, r

i

) has a nega-

tive effect on the competitor’s proﬁts. Fourth, the optimal advertising pol-

icy does not depend on the noisiness of the selling environment. Finally,

the results for the amount of advertising u

∗

1

are unambiguous and follow

the same intuition as in the symmetric case.

4.3. Characterization of the Evolution Path. We examine next the

market share paths analytically. Inserting the optimal controls into Eqs.

(7)–(8), we obtain

dx =

¸

β

1

ρ

2

1

/2c

1

+δ −x

β

1

ρ

2

1

/2c

1

+β

2

ρ

2

2

/2c

2

+2δ

¸

dt

+σ(x)dw, x(0) =x

0

, (33)

dy =

¸

β

2

ρ

2

2

/2c

2

+δ −y

β

1

ρ

2

1

/2c

1

+β

2

ρ

2

2

/2c

2

+2δ

¸

dt

−σ(1−y)dw, y(0) =1−x

0

. (34)

To keep the intermediate steps simple, we make use of the notation

A

1

≡β

1

ρ

2

1

/2c

1

+δ, A

2

≡β

2

ρ

2

2

/2c

2

+δ,

and derive the results only for ﬁrm 1. Rewriting (33) as the stochastic inte-

gral equation

x(t ) =x

0

+

t

0

(A

1

−x(s)(A

1

+A

2

))ds +

t

0

σ(x)dw, (35)

it is obvious that the mean evolution path is independent of the nature of

the stochastic disturbance. Speciﬁcally,

E[x(t )] =x

0

+

t

0

(A

1

−E[x(s)](A

1

+A

2

))ds. (36)

This can be expressed as an ordinary differential equation in E[x(t )], with

the initial condition E[x(0)] =x

0

, whose solution is given by

E[x(t )] =exp[−(A

1

+A

2

)t ]x

0

+(1−exp[−(A

1

+A

2

)t ])A

1

/(A

1

+A

2

). (37)

An analogous result is obtained for ﬁrm 2. The long run equilibrium mar-

ket shares. ( ¯ x, ¯ y) are obtained by taking the limit as t →∞ and are given

by

¯ x =A

1

/(A

1

+A

2

), ¯ y =A

2

/(A

1

+A

2

). (38)

174 JOTA: VOL. 123, NO. 1, OCTOBER 2004

Thus, the expected market shares converge to the form resembling the

attraction models commonly used in marketing. However, while an attrac-

tion model would rate the attractiveness of each ﬁrm based on its lower

cost, higher productivity of advertising, and higher advertising, it would

exclude exogenous market phenomena such as churn.

To further characterize the evolution path, we can calculate the vari-

ance of the market shares at each point in time. A speciﬁcation of the dis-

turbance function that satisﬁes (10) is required for this purpose. We use

σ(x)dw=σ

x(1−x)dw,

where σ is a positive constant.

An application of the It ˆ o formula to Eq. (33) provides the result,

d(x(t )

2

) =[2x(A

1

−x(A

1

+A

2

)) +σ

2

x(1−x)]dt

+2xσ

x(1−x)dw. (39)

Rewriting this as a stochastic integral, taking the expected value, and

rewriting as a differential equation, we get

(d/dt )E[x(t )

2

] =(2A

1

+σ

2

)E[x(t )] −(2A

1

+2A

2

+σ

2

)E[x(t )

2

]. (40)

Inserting the solution for E[x(t )] from (37), we obtain the following

ﬁrst-order linear differential equation in the second moment E[x(t )

2

]:

dE[x

2

]/dt +(2A

1

+2A

2

+σ

2

)E[x

2

]

=A

1

(2A

1

+σ

2

)/(A

1

+A

2

)

+exp[−(A

1

+A

2

)t ]

¸

(2A

1

+σ

2

)x

0

−A

1

(2A

1

+σ

2

)/(A

1

+A

2

)

¸

, (41)

with the initial condition

E[x(0)

2

] =x

2

0

.

The solution is

E[x(t )

2

] =x

2

0

e

−2(A

1

+A

2

+σ

2

/2)t

+

A

1

(A

1

+σ

2

/2)

(A

1

+A

2

+σ

2

/2)(A

1

+A

2

)

×[1−e

−2(A

1

+A

2

+σ

2

/2)t

]

+

e

−(A

1

+A

2

)t

−e

−2(A

1

+A

2

+σ

2

/2)t

A

1

+A

2

+σ

2

×

¸

2(A

1

+σ

2

/2)x

0

−

2A

1

(A

1

+σ

2

/2)

A

1

+A

2

. (42)

JOTA: VOL. 123, NO. 1, OCTOBER 2004 175

We can calculate the convergence of the second moment as the inﬂuence

of the initial condition disappears. That is,

lim

t →∞

E[x(t )

2

] =A

1

(A

1

+σ

2

/2)/(A

1

+A

2

+σ

2

/2)(A

1

+A

2

). (43)

Written in this form, it becomes clear that, when σ =0, the expression is

just ¯ x

2

, so that the variance is appropriately zero in the absence of sto-

chastic effect. More generally, when σ =0,

E[x(t )

2

] =(E[x(t )])

2

holds for all t . For σ >0, the variance is given by the formula E[x(t )

2

] −

(E[x(t )])

2

, which allows us to ﬁnd that, in the long run, the variance of

the market shares for both ﬁrms is given by

A

1

A

2

σ

2

/(2A

1

+2A

2

+σ

2

)(A

1

+A

2

)

2

. (44)

4.4. Illustration. Illustrative market shares may be obtained for

different parameter values. We choose the parameter values r = 0.05,

δ =0.01, symmetric margins m

1

=m

2

=1, asymmetric ﬁrm strengths R

1

=1,

R

2

=4, and an initial starting point at x(0) =0.5. In practice, a decision

calculus approach could be followed to obtain the parameter values. Using

Mathematica, we ﬁnd that the only real positive root for the quartic poly-

nomial is β

1

=0.264545 and the corresponding β

2

=0.43069. Finally, we

specify

σ(x)dw=σ

x(1−x)dw, with σ

2

=0.5.

The procedure described in Zwillinger (Ref. 28, pp. 702, Eq. 182.3) is used;

i.e., for an SDE

dx(t ) =a(x(t ))dt +b(x(t ))dw(t ),

the numerical approximation

x(t +) =x(t ) +a(x(t ))+b(x(t ))

√

ς(t )

can be used, where the {ς(t )} are i.i.d. standard normal random variables.

The time step =0.01 is used.

Figure 1 shows a sample path. One can see that the path hovers

around the mean. It never stays on the mean as it is continuously dis-

rupted due to the Brownian motion. We can calculate a conﬁdence inter-

val if the path is normally distributed around the mean. Then,

E[x(t )] ±1.96

E[x(t )

2

] −(E[x(t )])

2

176 JOTA: VOL. 123, NO. 1, OCTOBER 2004

Fig. 1. Market share trajectories given optimal advertising decisions.

provides the 95% conﬁdence interval for the market share path. While we

know that the distribution is not normal, nevertheless, as Fig. 2 shows, the

proposed conﬁdence interval does an adequate job of tracking the mar-

ket shares. Since the normal distribution is not bounded between zero and

one, the conﬁdence interval may exceed the minimum or maximum mar-

ket share as happened in this case, In Section 4.5, we obtain the equi-

librium distribution of the market share, which enables us to provide the

exact conﬁdence intervals for the equilibrium market shares.

This analysis has the value that it provides a diagnostic tool for man-

agement to handle market share ﬂuctuations. Whereas minor ﬂuctuations

within the conﬁdence bands may call for cursory examination, overstep-

ping the bands indicates the need for a detailed review. It may be indica-

tive of a shift in the underlying market parameters and hence it requires

a reevaluation of the advertising spending policies. Also, the market share

ﬂuctuations directly cause ﬂuctuations in advertising spending according

to Theorem 4.1; hence, one can simulate the advertising budget as well.

4.5. Probability Distribution of Market Shares. In Section 4.4, we

mentioned that the probability densities of the market shares are not nec-

essarily normally distributed. This raises the question of whether the den-

sity functions can be determined explicitly or at least approximated. We

devote this section to examining this issue.

JOTA: VOL. 123, NO. 1, OCTOBER 2004 177

Fig. 2. Market share for ﬁrm 1: Normal density 95% conﬁdence interval (dashed lines), and

equilibrium market share 95% conﬁdence interval (dotted lines).

An important property of the solution x(t ) of an It ˆ o stochastic differ-

ential equation,

dx(t ) =a(x, t )dt +b(x, t )dw(t ), x(s) =z,

is that it is a Markov process. The transition probability of this Markov

process has a density p(t, x; s, z) for going from market share z at time s

to market share x at time t >s, which satisﬁes the Fokker-Planck equation,

∂p/∂t +(∂/∂x)(ap) −(1/2)(∂

2

/∂x

2

)(b

2

p) =0,

p(t, x; t, z) =δ(x −z).

For our problem, we shall obtain ﬁrst and then attempt to solve the

Fokker-Planck equation. Firm 1’s stochastic differential equation from

(33) is

dx =(A

1

−(A

1

+A

2

)x)dt +σ

x(1−x)dw, x(0) =x

0

. (45)

The corresponding Fokker-Planck equation is given by

∂p/∂t +(∂/∂x)([A

1

−(A

1

+A

2

)x]p)

−(1/2)(∂

2

/∂x

2

)[σ

2

x(1−x)p] =0, (46)

178 JOTA: VOL. 123, NO. 1, OCTOBER 2004

which simpliﬁes to

(∂p/∂t ) +[σ

2

x(x −1)/2]∂

2

p/∂x

2

+([2σ

2

−(A

1

+A

2

)]x +A

1

−σ

2

)∂p/∂x +[σ

2

−(A

1

+A

2

)]p=0. (47)

This partial differential equation could not be solved explicitly. Neverthe-

less, we will attempt to ﬁnd the density of the steady-state market share

by lim

t →∞

p(t, x; s, z). Let f (x) denote this density, since it can be shown to

be independent of s, z, t .

To recapitulate, what we started off wanting to know was the density

p(t, x; 0, x

0

) of ﬁrm 1’s market share at time t , given that it starts at a

point x

0

at time zero. By looking for the long-run stationary probability

density of the market share, essentially we are willing to ignore the initial

transient part of the solution. For the density f (x), we can set ∂p/∂t =0

in (47) and obtain the second-order ordinary differential equation

[σ

2

x(x −1)/2] d

2

f /dx

2

+([2σ

2

−(A

1

+A

2

)]x +A

1

−σ

2

) df /dx

+[σ

2

−(A

1

+A

2

)]f =0. (48)

This is identiﬁable as a Gaussian hypergeometric equation with solution

(Ref. 29, pp. 234)

f (x) =x

(2A

1

/σ

2

−1)

(1−x)

(2A

2

/σ

2

−1)

×

C

1

+C

2

x

−2A

1

/σ

2

(1−x)

−2A

2

/σ

2

dx

. (49)

To determine the constants of integration, we can employ the follow-

ing two properties. First, the density should integrate to 1; second, the

expected value of the market share should be ¯ x, which we have already

calculated is equal to A

1

/(A

1

+A

2

). The result is given in the following

theorem.

Theorem 4.2. The density of the stationary distribution of a ﬁrm’s

market share is given by the beta density. For ﬁrm 1,

f (x) =

(2A

1

/σ

2

+2A

2

/σ

2

)

(2A

1

/σ

2

)(2A

2

/σ

2

)

x

2A

1

/σ

2

−1

(1−x)

2A

2

/σ

2

−1

, (50)

where

(s) =

∞

0

x

s−1

e

−x

dx, s >0,

JOTA: VOL. 123, NO. 1, OCTOBER 2004 179

is the gamma function. For ﬁrm 2, by symmetry, f (y) is obtained by

interchanging x with y and A

1

with A

2

in (50).

Proof. The density integrates to 1 by deﬁnition, while the mean of

the beta distribution is given by A

1

/(A

1

+A

2

). Thus, all the required con-

ditions are satisﬁed.

It can be veriﬁed that the variance of the beta distribution matches

the expression in (44) obtained by using the It ˆ o formula.

We now return to the illustrative example of Section 4.4. Inserting the

parameter values

A

1

=0.539, A

2

=3.46, σ

2

=0.5,

the beta density is

f (x) =292.39x

1.156

(1−x)

12.84

. (51)

We compute the 95% conﬁdence bounds to be l = 0.02 and m = 0.33.

These are sketched in Fig. 2. They provide a more accurate 95% conﬁ-

dence interval for the equilibrium market share of ﬁrm 1 than by assuming

a normal distribution. The conﬁdence intervals for ﬁrm 2 can be obtained

in a similar manner.

5. Conclusions

We examine a dynamic duopoly with stochastic disturbances and

employ closed-loop methods to solve the problem. The model is ana-

lyzed using stochastic differential game theory and explicit solutions are

obtained. The effects of several different parameters are discussed for sym-

metric and asymmetric ﬁrms.

The paper extends the work of Sethi (Ref. 15) to include competitive

advertising response and work of Sorger (Ref. 19) by including stochas-

tic analysis and a churn term in the dynamics that is consistent with the

original Vidale-Wolfe formulation and which ensures that, in the absence

of competitive advertising, the market shares will converge. The effect of

churn can be decomposed into two parts: one is to reduce competition by

making advertising less effective, hence causing a decrease in equilibrium

advertising; the other is to disproportionately reduce share of the higher

market share ﬁrm. Thus, higher churn beneﬁts a ﬁrm with low market

share, but has ambiguous effects for a higher share ﬁrm.

180 JOTA: VOL. 123, NO. 1, OCTOBER 2004

A simple rule describes the optimal advertising control, which is that

it should be proportional to the square root of the opponent market share

(Theorem 4.1). Sections 4.1 and 4.2 are devoted to determining the pro-

portionality constant, particularly its endogenous component β

i

, and to

obtaining an analytical expression for it. While it is given by a simple

expression when the ﬁrms are symmetric, it is not simple to state the pro-

portionality constant for asymmetric ﬁrms. An explicit formula is provided

in Appendix A, Section 6. Furthermore, the dependence of β

i

as well as

the amount of advertising is provided by means of comparative statics.

In Section 4.3, we characterize the evolution path by using stochastic

calculus to provide the mean and variance of the market shares. The for-

mer resembles an attraction model. In Section 4.5, we examine the prob-

ability distribution for the market shares by solving the Fokker-Planck

equation in the limiting case and showing that it is the beta distribution.

The fact that these commonly used forms for market share emerge endoge-

nously from the analysis validates additionally our model assumptions. An

illustration demonstrated the usability of the analysis in terms of tracking

the market shares (Section 4.4).

A limitation of the analysis is that it is restricted to duopoly compe-

tition, whereas many markets are characterized by more then two com-

petitors (Refs. 30–31). Future research should examine the possibility of

extending the present model to oligopoly markets. Extending the model

to incorporate decision variables such as price is also relevant. Finally,

the comparative statics presented in the paper represent hypotheses for

empirical testing which deserves further attention.

6. Appendix A: Proof of Uniqueness of Solution

We want to show that there exists a unique solution to the differential

game. This implies showing that there exists a unique β

1

>0 that satisﬁes

the quartic equation (29). We express (29) as

F(β

1

) ≡β

4

1

+κ

1

β

3

1

+κ

2

β

2

1

+κ

3

β

1

−κ

4

=0, (52)

where

κ

1

=2(W

1

+W

2

)/3R

1

, (53a)

κ

2

=

4m

2

R

2

−2m

1

R

1

−W

2

1

+2W

1

W

2

/3R

2

1

, (53b)

κ

3

=2m

1

(W

1

−W

2

)/3R

2

1

, (53c)

κ

4

=m

2

1

/3R

2

1

. (53d)

JOTA: VOL. 123, NO. 1, OCTOBER 2004 181

Every quartic equation has four roots. Excluding the fortuitous cases,

where two or more roots are equal, the following results are easily

observed:

(A1) When β

1

→±∞, F(β

1

) →∞; when β

1

= 0, F(β

1

) < 0. Since

F(β

1

) is differentiable, it is continuous. Thus, it must cross the

x-axis at least twice ensuring at least one positive real root and

one negative real root. If there are only two real roots, one will

be positive and one negative. If all four roots are real, they will

be either three positive and one negative or three negative and

one positive.

(A2) In the case of three real positive roots, ordering them from

the smallest to the largest, the slope at the second largest root

must be negative. To see this, write

F(β

1

) =(β

1

−β

1

(1))(β

1

−β

1

(2))(β

1

−β

1

(3))(β

1

−β

1

(4)),

where

β

1

(1) <0, β

1

(2) >0, β

1

(3) >β

1

(2), β

1

(4) >β

1

(3)

are the four roots. Then, the slope at β

1

(3),

F

(β

1

)|

β

1

=β

1

(3)

=(β

1

(3) −β

1

(1))(β

1

(3) −β

1

(2))(β

1

(3) −β

1

(4)),

is negative.

(A3) We can calculate the slope F

(β

1

) directly from (52) and eval-

uate it at any positive real root. The result is

F

(β

1

)|

β

1

=F

−1

(0)

=2

¸

3R

2

1

β

4

1

+R

1

(W

1

+W

2

)β

3

1

+m

1

W

2

β

1

+m

1

(m

1

−W

1

β

1

)

¸

/β

1

>0. (54)

It follows from points (A2) and (A3) above that there is only

one real positive root and hence a unique solution to the

differential game. To obtain an explicit solution, we utilize the

182 JOTA: VOL. 123, NO. 1, OCTOBER 2004

Mathematica 4.1 software to generate four solutions to (52):

β

1

(1) =−κ

1

/4−g/2−0.5

×

3κ

2

1

/4−2κ

2

−g

2

+

κ

3

1

−4κ

1

κ

2

+8κ

3

/4g, (55)

β

1

(2) =−κ

1

/4−g/2+0.5

×

3κ

2

1

/4−2κ

2

−g

2

+

κ

3

1

−4κ

1

κ

2

+8κ

3

/4g, (56)

β

1

(3) =−κ

1

/4+g/2−0.5

×

3κ

2

1

/4−2κ

2

−g

2

−

κ

3

1

−4κ

1

κ

2

+8κ

3

/4g, (57)

β

1

(4) =−κ

1

/4+g/2+0.5

×

3κ

2

1

/4−2κ

2

−g

2

−

κ

3

1

−4κ

1

κ

2

+8κ

3

/4g, (58)

where the two intermediate terms g and h are deﬁned as

g ≡

κ

2

1

/4−2κ

2

/3+2

1/3

κ

2

2

−3κ

1

κ

3

−12κ

4

/3h+h/32

1/3

, (59)

h≡

¸

2κ

3

2

−9κ

1

κ

2

κ

3

+27κ

2

3

−27κ

2

1

κ

4

+72κ

2

κ

4

+

−4

κ

2

2

−3κ

1

κ

3

−12κ

4

3

+

2κ

3

2

−9κ

1

κ

2

κ

3

+27κ

2

3

−27κ

2

1

κ

4

+72κ

2

κ

4

2

¸

1/3

. (60)

We pick β

1

as the only real positive solution out of the four roots, i.e.,

β

1

=β

1

(i

∗

), where i

∗

={i ∈{1, 2, 3, 4}|β

1

(i) >0}. (61)

While i

∗

may depend on the data, there will only be one i

∗

in every case.

7. Appendix B: Outline of Proof of Comparative Statics for Table 2

(B1) To obtain comparative static results for β

i

, we deﬁne

G

1

(β

1

, β

2

) ≡m

1

−β

2

1

ρ

2

1

/4c

1

−β

1

β

2

ρ

2

2

/2c

2

−β

1

(r

1

+2δ) =0, (62)

G

2

(β

1

, β

2

) ≡m

2

−β

2

2

ρ

2

2

/4c

2

−β

1

β

2

ρ

2

1

/2c

1

−β

2

(r

2

+2δ) =0. (63)

JOTA: VOL. 123, NO. 1, OCTOBER 2004 183

Then, for any parameter θ, the implicit function theorem may

be written as

∂β

1

∂θ

∂β

2

∂θ

=−(1/)

×

¸

¸

¸

−

β

2

ρ

2

2

2c

2

+

β

1

ρ

2

1

2c

1

+r

2

+2δ

β

1

ρ

2

2

2c

2

β

2

ρ

2

1

2c

1

−

β

1

ρ

2

1

2c

1

+

β

2

ρ

2

2

2c

2

+r

1

+2δ

×

∂G

1

∂θ

∂G

2

∂θ

, (64)

where >0. The comparative static results for changes in the

parameters c

2

, m

1

, m

2

, r

1

, r

2

, ρ

2

, δ follow in a straightforward

manner. However, the cases for c

1

and ρ

1

could not be signed.

(B2) For comparative statics for u

∗

1

, we insert

u

∗

1

=β

1

ρ

1

√

1−x/2c

1

,

u

∗

2

=β

2

ρ

2

√

x/2c

1

into (62)–(63) to obtain G

1

(u

∗

1

, u

∗

2

) and G

2

(u

∗

1

, u

∗

2

). Then, the

implicit function theorem can be written as

∂u

∗

1

/∂θ

∂u

∗

2

/∂θ

=−(1/)

×

¸

¸

¸

−

2c

2

u

2

x

+

2c

2

u

1

ρ

1

ρ

2

√

1−x

√

x

+

2c

2

(r

2

+2δ)

ρ

2

√

x

2c

1

u

1

ρ

2

ρ

1

√

1−x

√

x

2c

2

u

2

ρ

1

ρ

2

√

1−x

√

x

−

2c

1

u

1

1−x

+

2c

1

u

2

ρ

2

ρ

1

√

1−x

√

x

+

2c

1

(r

1

+2δ)

ρ

1

√

1−x

×

∂G

1

/∂θ

∂G

2

/∂θ

,

(65)

where it can be shown that > 0. The calculations for the

results reported in Table 2 follow in a straightforward manner.

(B3) For α

1

, we note from Eq. (30) that, in many cases, α

1

will have

the same comparative statics as β

1

. These relationships are as

follows:

∂α

1

/∂c

2

>0, ∂α

1

/∂m

1

>0, ∂α

1

/∂m

2

<0, ∂α

1

/∂r

1

<0,

∂α

1

/∂r

2

>0, ∂α

1

/∂ρ

2

<0. (66)

184 JOTA: VOL. 123, NO. 1, OCTOBER 2004

the cases for c

1

, ρ

1

, δ are unclear.

(B4) The unambiguous results for V

1

occur when comparative stat-

ics for α

1

and β

1

are in the same direction, which is true for

all parameters except c

1

, ρ

1

, δ.

References

1. Vidale, M. L., and Wolfe, H. B., An Operations Research Study of Sales

Response to Advertising, Operations Research, Vol. 5, pp. 370–381, 1957.

2. Sethi, S. P., Dynamic Optimal Control Models in Advertising: A Survey, SIAM

Review, Vol. 19, pp. 685–725, 1977.

3. Feichtinger, G., Hartl, R. F., and Sethi, S. P., Dynamic Optimal Control

Models in Advertising: Recent Developments, Management Science, Vol. 40,

pp. 195–226, 1994.

4. CMR/TNS Media Intelligence, US Advertising Market Shows Healthy Growth:

Spending Up 4.2% in 2002; see http://www.tnsmi-cmr.com/news/2003/031003.html

(March 10, 2003).

5. Patti, C. H., and Blasko, V. J., Budgeting Practices of Big Advertisers, Jour-

nal of Advertising Research, Vol. 21, pp. 23–29, 1981.

6. Blasko, V. J., and Patti, C. H., The Advertising Budgeting Practices of Indus-

trial Marketers, Journal of Marketing, Vol. 48, pp. 104–110, 1984.

7. Aaker, D., and Carman, J. M., Are You Overadvertising? Journal of Advertis-

ing Research, Vol. 22, pp. 57–70, 1982.

8. Lodish, L.M., Abraham, M., Kalmenson, S., Livelsberger, J., Lubetkin, B.,

Richardson, B., and Stevens, M. E., How TV Advertising Works: A Meta-

Analysis of 389 Real World Split Cable TV Advertising Experiments, Journal

of Marketing Research, Vol. 32, pp. 125–139, 1995.

9. Chintagunta, P. K., and Vilcassim, N., An Empirical Investigation of Adver-

tising Strategies in a Dynamic Duopoly, Management Science, Vol. 38, pp.

1230–1224, 1992.

10. Erickson, G. M., Empirical Analysis of Closed-Loop Duopoly Advertising

Strategies, Management Science, Vol. 38, pp. 1732–1749, 1992.

11. Fruchter, G., and Kalish, S., Closed-Loop Advertising Strategies in a Duop-

oly, Management Science, Vol. 43, pp. 54–63, 1997.

12. Bain, J. S., Barriers to New Competition, Harvard University Press, Cam-

bridge, Massachusetts, 1956.

13. Nelson, P., Advertising as Information, Journal of Political Economy, Vol. 82,

pp. 729–754, 1974.

14. McAlister, L., and Pessemier, E., Variety Seeking Behavior: An Interdisciplin-

ary Review, Journal of Consumer Research, Vol. 9, pp. 311–323, 1982.

15. Sethi, S. P., Deterministic and Stochastic Optimization of a Dynamic Advertising

Model, Optimal Control Applications and Methods, Vol. 4, pp. 179–184, 1983.

JOTA: VOL. 123, NO. 1, OCTOBER 2004 185

16. Deal, K., Sethi, S. P., and Thompson, G. L., A Bilinear-Quadratic Differen-

tial Game in Advertising, Control Theory in Mathematical Economics, Edited by

P. T. Lui and J. G. Sutinen, Manuel Dekker, New York, NY, pp. 91–109, 1979.

17. Chintagunta, P. K., and Jain, D. C., Empirical Analysis of a Dynamic Duop-

oly Model of Competition, Journal of Economics and Management Strategy,

Vol. 4, pp. 109–131, 1995.

18. Horsky, D., and Mate, K., Dynamic Advertising Strategies of Competing

Durable Good Producers, Marketing Science, Vol. 7, pp. 356–367, 1988.

19. Sorger, G., Competitive Dynamic Advertising: A Modiﬁcation of the Case

Game, Journal of Economics Dynamics and Control, Vol. 13, pp. 55–80, 1989.

20. Little, J. D. C., Aggregate Advertising Models: The State of the Art, Opera-

tions Research, Vol. 27, pp. 629–667, 1979.

21. Sethi, S. P., Optimal Control of the Vidale-Wolfe Advertising Model, Opera-

tions Research, Vol. 21, pp. 998–1013, 1973.

22. Erickson, G. M., Dynamic Models of Advertising Competition, Kluwer, Nor-

well, Massachusetts, 2003.

23. Dockner, E. J., Jorgensen, S., Long, N. V., and Sorger, G., Differential

Games in Economics and Management Science, Cambridge University Press,

Cambridge, UK, 2000.

24. Kimball, G. E., Some Industrial Application of Military Operations Research

Methods, Operations Research, Vol. 5, pp. 201–204, 1957.

25. Gihman, I. I., and Skorohod, A. V., Stochastic Differential Equations,

Springer-Verlag, New York, NY, 1972.

26. Gould, J. P., Diffusion Processes and Optimal Advertising Policy, Microeco-

nomic Foundations of Employment and Inﬂation Theory, Edited by E. S.

Phelps et al., W. W. Norton, New York, NY, pp. 338–368, 1970.

27. Sethi, S. P., and Thompson, G. L., Optimal Control Theory: Applications to

Management Science and Economics, Kluwer, Norwell, Massachusetts, 2000.

28. Zwillinger, D., Handbook of Differential Equations, Academic Press, San

Diego, California, 1998.

29. Polyanin, A. D., and Zaitsev, V. F., Handbook of Exact Solutions for Ordi-

nary Differential Equations, Chapman and Hall, CRC, Boca Raton, Florida,

2003.

30. Fruchter, G., Oligopoly Advertising Strategies with Market Expansion, Opti-

mal Control Applications and Methods, Vol. 20, pp. 199–211, 1999.

31. Erickson, G. M., Advertising Strategies in a Dynamic Oligopoly, Journal of

Marketing Research, Vol. 32, pp. 233–237, 1995.

- Handout EightUploaded byCassie Desilva
- Stability Assessment of a Critical Rock SlopeUploaded byhemubc
- CFA Exam Result ExplanationUploaded byGart Desa
- Risk Reliability Summary and TablesUploaded byJohn smith
- Efficiency of the Philippine Stock MarketUploaded byfirebirdshockwave
- 39 3 Sums n Diffs Rndm VarsUploaded byEbookcraze
- Statistics for business 5Uploaded byHai Le
- Condicion de Mantto y Descripcion Del SistemaUploaded byEli Pale
- IntroductionUploaded bySameera Calou
- Markov Chain ExercisesUploaded byJason Wu
- Statistics and Data With RUploaded byabrax65
- Chapter_11 Sampling Distributions and EstimationUploaded byJacob Bains
- Thoughts About EstimationUploaded byRoger Siegenthaler
- 3-2GradingUploaded byJohny Villanueva
- PertUploaded byvaithy2011
- Wikipedia - Standar DeviasiUploaded byDik Sadja
- 601Example Statistic ProjectUploaded byEng Stephen Arende
- Chapter 8 and 9 Stats ReviewUploaded byAbby
- MKT RES 5-6Uploaded bySyed Sazzad Ali
- Formula Sheet Non-cumulative Final F2011Uploaded byBradley Chan
- Handout7Uploaded byahmed22gouda22
- CH17 Mock TestUploaded bynageswara_mutyala
- Probability, Statistics and Stochastic ProcessesUploaded byZhang Lizi
- MIL-STD-781C and confidence intervals on MTBF.pdfUploaded byrab__bit
- a069966.pdfUploaded bymangyan
- 5 BiomechanicsUploaded byPande Putu Agus Hendrawan
- statsprojectUploaded byapi-358676978
- Introduction to Statistics With SASUploaded byRaj Malhotra
- Week8AssessmentUploaded byJames Bowman
- Nasa/Tm—2005 213958Uploaded byjr-parshanth

- leadership preferences in japan 2007Uploaded bykashif salman
- organisational knowledge leadership- grounded theory approach-2007Uploaded bykashif salman
- sunflowerUploaded bykashif salman
- challenges and prospects of hrm practices in developing countriesUploaded bykashif salman
- organisational support, employee development and commitmentUploaded bykashif salman
- leadership theoryUploaded bykashif salman
- gender and ownership in uk small firmsUploaded bykashif salman
- leadership of a culture change processUploaded bykashif salman
- Introduction to Microsoft OfficeUploaded bykashif salman
- women enterpreneurs- management skills and business problemsUploaded bykashif salman
- Children Attitude Toward TVC in PakUploaded bykashif salman
- imapct of selected hr parctices on percieved emplyee performanceUploaded bykashif salman
- transformational leadershipUploaded bykashif salman
- organisational commitment among pakistani university teachersUploaded bykashif salman
- uk university articleUploaded bykashif salman
- article12Uploaded bykashif salman
- markusUploaded bykashif salman
- appraisal2Uploaded bykashif salman
- holy placesUploaded bykashif salman
- article4Uploaded bykashif salman
- afshan 4Uploaded bykashif salman
- max planckUploaded bykashif salman
- advertising bansUploaded bykashif salman
- appraisalUploaded bykashif salman
- advertising under varticle producat differenateUploaded bykashif salman
- appraisal employee developmentUploaded bykashif salman
- appraisal1Uploaded bykashif salman
- advertising create a goodwill of producateUploaded bykashif salman
- advertising planningUploaded bykashif salman
- alien definitionUploaded bykashif salman