ARIMA Modeling

:
B-J Procedure

A k Chauhan v_akc@rediffmail.com 9811216905

Structural models multivariate in nature, and Y = F (movements in current or past values of other (explanatory) variables). Univariate time series models Yt = F (own past values, Current & past values of an error term) useful when a structural model is inappropriate. when other variables are not observable or not measurable Explanatory variables are measured at a lower frequency of observation than yt. For eg, yt might be a series of daily stock returns, where possible explanatory variables could be macroeconomic indicators that are available monthly.

1ut-1 + 2 ut-2 + .. yt = µ + ut + is a MA(q) model.Moving average processes: Let ut be a white noise process with zero mean & constant variance Then 2. + q ut-q ..

depends upon only the values that the variable took in previous periods plus an error term. denoted as AR(p). can be expressed as Yt = u + 1 yt-1 + 2 yt-2 + .Autoregressive processes: An AR model is one where the current value of a variable y. + p yt-p + ut ‡ where ut is a white noise disturbance term.. An AR model of order p. The expression of AR(p) model can be written more compactly using sigma notation .

+ p yt-p + q ut-q + ut .q) processes obtained by combining the AR(p) & MA(q) models. q)m model ..+ Is a ARMA (p.. The model could be written Yt = u + 1 yt-1 + 2 yt-2 + 1 ut-1 + 2 ut-2 + . States that the current value of some series y depends linearly on its own previous values plus a combination of current and previous values of a white noise error term.ARMA (p.

.An AR process has: a geometrically decaying ACF a number of non-zero points of PACF = AR order. A MA process has: number of non-zero points of ACF = MA order a geometrically decaying PACF. A combination ARMA process has: a geometrically decaying ACF a geometrically decaying PACF.

Process of ARIMA Modeling ‡ ‡ ‡ ‡ Identification Stage Estimation Stage Diagnostic Checking Forecasting .

. Detrend and Differentiate the series if required to obtain a stationary series. p and q as small as possible. In order to achieve parsimony keep the number of AR and MA lag orders. provided the model still satisfactorily forecast the series. It is better to used mixed ARMA (p. Identify the undrlaying AR (p) and MA (q) processes in the stationary series based on the behaviour of the correlogram.Identification ‡ ‡ ‡ ‡ ‡ First identify the order of integration on the basis of visual inspection of time series plot. correlogram and formal unit root testing. q) model rather than pure AR (p) or MA (q) model.

known as Max likelihood.Estimation involves estimation of parameters of model specified in step 1. . This can be done using least squares or another technique. depending on the model.

Suggested methods are: Overfitting -. Residual diagnostics. any extra terms added to ARMA model would be insignificant. Information Criteria: SBIC. AIC HQ should have least values. Residual should be a white noise series. .fitting a larger model than that required. If the model specified at step 1 is adequate.Model checking determining whether the model specified and estimated is adequate.

.

.

.

Negative MA Coeff .

.

.

Negative MA Coeff .

.

.

.

.Estimation Stage ‡ A pure AR model can be estimated using OLS estimator but if the model also includes MA terms ML estimator is used.

Diagnostic Checking .

How good the model fits the data? How reliable will be the prediction based on it? ‡ Use some indicators of goodness of fit such as R2 and ANOVA. .

Q and LB statistics of residuals is useful for detecting serial correlation.Is the model sufficiently specified or it requires including some more AR or MA terms? Examine the residuals. Use LR test for AR in residuals. Their correlogram can give a clue if residuals are pure white noise or at least free from serial correlation. .

Does the model include some unnecessary parameters (AR or MA terms) which can be dropped out without a significant loss in predictive efficiency of the model? ‡ t-statistic of individual coefficient can give an initial (but not conclusive) clue. . .

‡ In an ARIMA (p. and ± at original integrated time series. the forecasting can be made at two different levels ± at the level of differenced stationary time series. .q) model where d>0.Forecasting ‡ In ARMA (p.d. q) the prediction equation is simply a linear equation that refers to past values of original time series and past values of the errors.

Sign up to vote on this title
UsefulNot useful