Distinction Between Chance Causes Forecasting is a technique that uses abulated Value = 1.
ulated Value = 1.72( at 5% level of n(Σfi) = 230 , Median = 46 From the
/Any one chance causes results in historical data as inputs to make significance with 21 degrees of table, Σfi = 150 + x + y , Here, n =
only a minute amount of variation informed estimates that are freedom) as shown in the table 230 n/2 = 115 Given , Median = 46,
/Some typical chance causes of predictive in determining the direct below. which belongs to the class 40 - 50 .
variation are (a) Slight variations in ion of future trends. Businesses So the Median class is 40 - 50 . Here,
raw material (though within the utilize forecasting to determine how l = 40 , f = 65, c.f = (42 + x) , h = 10
specifications) (6) Slight vibration of to allocate their budgets or plan for MEDIAN = l + [(n/2 - cf )/f ] ×h
The mean weakly sales of soap bar in
machine (c) Lack of human perfectio anticipated expenses for an upcomi 46 = 40 + [(115 - (42+x))/65] × 10
department stores was 146.3bars
n in reading instrum ents/As a practi ng period of time/A forecast refers store.after an advertising campaign the 46 - 40 = [(115 - (42+x))/65] × 10
cal matter chance variation cannot to a calculation or an estimation mean weakly sales in 22 stores for a 6 = [(115 - 42 - x)/65] × 10 /6 =[ (73 -
economically be eliminated from a which uses data from previous typical week increased to 153.7 and x)/65]×10 /(6× 65)/10 = 73 - x
process/ Assignable CausesConsists events, combined with recent trends showed a S.D. 17.2 was the advertising (3 × 65)/5 = 73 - x ;3 × 13 = 73 - x
of one or just a few individual to come up a future event outcome. campagin successfulThe provided 39 = 73 - x ;x = 73 - 39 ;x = 34 ;Σfi =
causes /Any one assignable cause On the other hand, a prediction is an sample meanis \ 150 + x + y ;230 = 150 + 34 + y [Σfi
can result in a large amount of actual act of indicating that somethi bar{x}=153.7xˉ=153. 7 and the = 230 , x = 34] ;230 = 184 + y ;y =
variation/Some typical assignable ng will happen in the future with or sample standard deviation 230 - 184 ;y = 46 ;Hence, the
causes of variation are : Batch of without prior information/ Forecas is s=17.2,s=17.2, and the sample size missing frequencies be x = 34 and y
defective raw material ting models are one of the many is n=22.n=22.The following null and = 76 .
Faulty set up Untrained operator tools businesses use to predict outc alternative hypotheses need to be (ii) MEAN: MEAN = A + h ×(Σfiui
/The presence of assignable omes regarding sales, supply and tested:H_0: \mu=146.3H0:μ=146.3 /Σfi) From the table, Σfiui = 20 , Σfi
variation can be detected and action demand, consumer behavior and H_1: \mu>146.3H1:μ>146.3 This = 230 Let the assumed mean, A =
to elimin ate the causes is usually more. These models are especially corresponds to a right-tailed test, 45, h = 10 Mean = 45 + 10(20/230)
justified/ beneficial in the field of sales and for which a t-test for one mean, with = 45 + 20/23 = 45 + 0.87 = 45.87
a shopkeeper claims that the avera marketing. There are several unknown population standard Hence, the Mean is 45.87 .
ge life of CFL bulb is 1600 hours.to forecast ting methods businesses deviation will be used. Based on the
check this claim a researcher takes use that provide varying degrees of information provided, the significan
a sample of 100 CFL bulbs of tha informati on. From the simple to the ce level is \alpha=0.05, df=n-1=22-
same make randomly and finds complex, the appeal of using 1=21α=0.05,df=n−1=22−1=21 andth
mean lifetime of 1570 hours with forecasting models comes from e critical value for a right-tailed test
standard daviation of 120 hours.is having a visual reference of is t_c=1.721tc=1.721 The rejection
tha claim acceptable at 5% level of expected outcomes region for this right-tailedtestis R =\
significance Suppose 2% of the items made in a {t:t>1.721\}.R={t:t>1.721}.The t-
factory are defective. Find the statistic is computed as follows:t ={\
n=100/xˉ=1570 σ=120 H0:μ= 1600
probability that there are (a) 3 defective bar{x}-\mu_0 \over s/\sqrt{n}} One-Way ANOVA A one-way layout
H1:μ=1600 Test-statistic: items in a sample of 100, and (b) no
={153.7-146.3 \over17.2/\sqrt{22}} \ consists of a single factor with
defective item in a sample of 2% of
approx2.018t=s/nxˉ−μ0=17.2/22153 several levels and multiple
the items of a factory are .7−146.3≈2.018Since it is observed observations at each level. With this
=−2.5α=0.05 Two-tailed test. Reject defective.=> p = probability of that t=2.018>1.7221=t_c,t=2.018>1. kind of layout we can calculate the
H0 if Z ≤1.96 or Z≥1.96.We are able defective = 0.02q = probability of 7221=tc, then it is mean of the observations within
to reject H0. The popula tion mean non-defective = 1 - 0.02 = 0.98P(x) concludedthat the null hypothesis is each level of our factor. The
lifetime of bulbs is not equal to 1600 = ⁿCₓpˣqⁿ⁻ˣ(i) 3 defective items n = rejected. There fore, there is enough residuals will tell us about the
hours at 0.05 significa nce level./ 100 x = 2=> P(3) = ¹⁰⁰C₃(0.02)³ evidence to claim that the variation within each level. We can
The following contingency table (0.98)⁹⁷= 0.1823Probability 3 defe population mean is greater than also average the means of each level
presents the analysis of 300 ctive items in 100 = 0.1823 n=50 146.3, at the 0.05 signi ficance to obtain a grand mean. It is easy to
persons according to hair colour 0 defective items P(0) level.Using the P-value model all of this with an equation of
and eye colour: Hair Colour Eye =⁵⁰C₀(0.02 )⁰(0.98)⁵⁰ = 0.36417 approach: t=2.018, df=22-1, \alpha= the form:
Colour 0.05, one tailedt=2.018,df=22−1,
Probabilit y 0 defective items in 50 =
H0 : Hair and eye colour are 0.364 α=0.05,one−tailed The p-value
independentH1 : Hair and eye The mean weakly sales of soap bars in is p=0.028274,p=0.028274, andsince
The equation indicates that the jth
colour are associateχ2=∑\sum∑ri=1 departmental stores were 146.3 bars p=0.028274<0.05=\
data value, from level i, is the sum of
∑\sum∑ cj=1=(Oij-Eij)2/Eij per store. After an advertising campaign alpha,p=0.028274<0.05=α, it is
three components: the common
Eij=RiCj/N, Ri - sum of i-th row, Cj - the mean weekly sales in 22 stores for a concluded that the null hypothesis is
typical week increased to 153.7 and valu e (grand mean), the level effect
sum of j-th column, N - total sample rejected. Therefor e, there is enough
showed a SD of 17.2. Was the (the deviation of each level mean
sizeE11=(80*120)/ 300=32, evidence to claim that the
advertising campaign successful? t from the grand mean), and the
(O11E11)2=4/E12=(80*60 )/300=16, population mean is greater than
test for single mean is given by the residual
(O12E12)2=36/E13=(80* 146.3, at the 0.05 significance
120)/300=32, (O13-E13)2=64/E21 level.Hence the advertising campa
=(100*120)/300 =40, (O21-E21)2=0 ign was successful in promo ting
formula Where sales An incomplete distribution is
E22=(100*60)/300=20, (O22-E22)2=
0E23=(100*120)/300=40, (O23- x‾\overline{x}x is the sample mean, given below :Variable: 10−20 20−30
E23)2=0E31=(120*120)/300=48, µ is the population mean, s is the SD 30−40
(O31-E31)2=4E32=(120*60)/300=24, and n is the number of observations 40−50 50−60 60−70 70−80
Frequency: 12 30 – 65 – 25 18You are
(O32-E32)2=36 E33=(120*120)/300=
given that the median value is 46 and
48, (O33-E33)2=64χ2=7.29degrees s=17.2, n=22 n = 22n=22 in the the total number of items is 230. (i)
of freedom: (r-1)(c-1)=4the critical above formula we get the calculated Using the median formula fill up
value of chi square with 4df at 5% Using ANOVA software or the tech
t value is missing frequen cies. (ii) Calculate
level significance: 9.488χ2<critical niques of the value-splitting exampl
the AM of the completed
value, so we accept H0 ,so hair e, we summarize the data in an
distribution. Let the missi ng
colour is independent of eye colour/ ANOVA table as follows:
frequencies be x and y. Given :
The die is unbiasedAlternative
Hypothesis (HA): The die is not
unbiased.Step 2 : Test Statistics
On the hypothesis that the die is
Time Series The Series of data unbiased we should expect the
points recorded over a specified frequency of each number to be
period of time is called Time-series 132/6=22∴Xcal2=∑(O−E)2E=198/22=
data. Time-series analysis is a 8.91Step 3:L.O.S (∝) = 0.05Degree of
technique for analyzing time series freedom = n-1 = 6-1 =5Step 4 :∴
data and extract meaningful Critical value (X∝)2=11.0705Step 5 :
statistical infor mation and DecisionSince Xcal2< (X∝)2
characteristics of the data. One of HO is accepted.∴ The die is
the major objectives of the analysis unbiased.
is to forecast future value.
Extrapolation is involved when
forecasting with the time series
analysis which is extremely complex.
But, the forecasted value along with
the estimation of uncertainty assoc
iated with that can make the result
extremely valuable.time series anal
ysis is a statistical method to analyse
the past data within a given duration
of time to forecast the future. It
comprises of ordered sequence of
data at equally spaced interval.The
first step is to perform the explor
atory analysis which is carried out by
plotting a line chart of the count of
passengers against time. Figure 1
shows the count of passenger on y-
axis and time on x-axis where each
interval can be considered as a year
Systematic sampling is a probability
sampling method in which a random
sample, with a fixed periodic
interval, is selected from a larger
population. The fixed periodic
interval, called the sampling
interval, is calculated by dividing the
population size by the desired
sample size Advantages of
Systematic Sampling It is simple
and convenient to use.Researchers
can create, analyze, and conduct
samples easily when using this
method because of its
structure/There isn’t a need to
number each member of a
sample.Researchers can represent
an entire population quickly and
easily when using systematic sampl
ing/The created samples are based
on precision.The samples that get
created from systematic sampling
have a higher level of precision than
other randomized methods Disadv
antages of Systematic Sampling
This process requires a close approxi
mation of a population/Some
popula tions can detect the pattern
of sampl ing/It creates a fractional
chance of selection/A high risk of
data manip ulation exists A die is
rolled 1200 times with the following
results : No. that comes up : 1 2 3 4 5 6
Frequency : 195 289 202 242 163 109
Test, if the die. is unbiased at 5% level
of significance. (Given that x 0.05 (5) =
11.07)Step 1 :Null hypothesis : (H0)