Scribd
Upload
5 views3 pages

Assignment 3

The document analyzes autocorrelation in an AR(1) regression model for housing services, concluding that the model's restrictions are not valid, leading to the adoption of an unrestricted ADL(1,1) model. The analysis indicates that while the lagged dependent variable and current income significantly impact housing demand, lagged income and price variables do not, suggesting a partial adjustment model is more appropriate. Additionally, a separate analysis of the NIFTY 50 market proxy indicates weak form efficiency, as no autocorrelation was found in the returns.

Uploaded by

akhil cv
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views3 pages

Assignment 3

The document analyzes autocorrelation in an AR(1) regression model for housing services, concluding that the model's restrictions are not valid, leading to the adoption of an unrestricted ADL(1,1) model. The analysis indicates that while the lagged dependent variable and current income significantly impact housing demand, lagged income and price variables do not, suggesting a partial adjustment model is more appropriate. Additionally, a separate analysis of the NIFTY 50 market proxy indicates weak form efficiency, as no autocorrelation was found in the returns.

Uploaded by

akhil cv
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ASSIGNMENT No 3:

PART - 1
Autocorrelation in AR(1) Regression for Housing Services

Autocorrelation in time series regression, indicates whether past values of a variable are correlated
with its present values. It is essential to test for autocorrelation to ensure that model estimates are
unbiased and efficient. In this assignment, we analyze the presence of autocorrelation in an
autoregressive model of order one (AR(1)) for housing services. The output from the regression
analysis is evaluated using standard tests, including the Durbin-Watson statistic and the h-statistic, to
determine the adequacy of the model specification. Further, we assess whether the restrictions
imposed by the AR(1) model are valid or whether an alternative unrestricted model is preferable.

Regression Analysis
The unrestricted model under consideration is estimated using OLS on a sample spanning from 1960
to 1994. The dependent variable is LGHOUS, and the key explanatory variables include:
• LGDPI: Represents the level of income available for households to spend on housing services.
• LGPRHOUS: Captures the effect of house price variations on housing services demand.
• Lagged dependent and independent variables: To account for potential dynamic
relationships in housing services over time.
The regression results indicate that the lagged dependent variable (LGHOUS(-1)) is highly significant,
with a coefficient of 0.7267 and p value close to 0. The variable LGDPI is also significant, suggesting
that current income has a direct impact on housing services demand. However, its lag (LGDPI(-1)) is
not significant. Similarly, the lag of LGPRHOUS(-1) is not statistically significant.

Autocorrelation Testing
Before proceeding with model specification tests, autocorrelation is checked, using the following:
1. Durbin-Watson Statistic: The reported value of 1.517562 suggests potential positive
autocorrelation. However, it falls within an intermediate range where no definite conclusion
can be drawn.
2. h-Statistic: This test is particularly useful in large sample sizes when the Durbin-Watson
statistic is inconclusive. The h-statistic is given by:

Since this value is well below the critical threshold of 1.96, we fail to reject the null hypothesis of no
autocorrelation. Thus, there is no strong evidence of autocorrelation affecting the regression results.
Checking whether the coefficients appear to satisfy the restrictions implicit in the AR(1) model.
Minus the product of the lagged dependent variable and the income elasticity is 0.7267×0.3014 = –
0.22. The coefficient of lagged income is numerically much lower than this. Minus the product of the
lagged dependent variable and the price elasticity is –0.7267 × –0.1924 = 0.14, which is identical to
the coefficient of lagged price, to two decimal places. Hence the restriction for the price side of the
model appears to be satisfied, but that for the income side does not. The common factor test
confirms this preliminary observation. The residual sum of squares has fallen to 0.000906. The test
statistic is 35 log(0.001360/0.000906) = 14.22. The critical value of chi squared at the 0.1 percent
level with two degrees of freedom is 13.82, so we reject the restrictions implicit in the AR(1) model
and conclude that we should use the unrestricted ADL(1,1) model instead. We note that the lagged
income and price variables in the unrestricted model do not have significant coefficients, so we
consider dropping them and arrive at the partial adjustment model specification already considered
above. As we saw, the h statistic is 0.86, and we conclude that this is a satisfactory specification

Model Selection and Final Considerations


Given the rejection of AR(1) model restrictions, we consider the unrestricted ADL(1,1) model, which
allows for more flexibility in capturing dynamic relationships. The insignificance of lagged income and
price variables in the unrestricted model suggests that they can be dropped to simplify the model.
This leads to a partial adjustment model, which accounts for short-run adjustments while
maintaining a stable long-run relationship.

The final check involves reassessing autocorrelation in the simplified model. The h-statistic in this
case is 0.86, which is within an acceptable range, confirming that the model does not suffer from
significant autocorrelation.

Conclusion
The analysis demonstrates that the AR(1) model does not provide an adequate specification for
housing services data due to its restrictive assumptions. The unrestricted ADL(1,1) model offers a
better fit, as confirmed by the common factor test. However, further refinement suggests that a
partial adjustment model may be the most appropriate specification. The absence of significant
autocorrelation in the final model indicates that it provides a reliable representation of housing
services behaviour.

PART-2

Check whether market is weak form efficient using the stock considered in Assignment 2:
Regressing NIFTY 50 (market proxy) with its lag value
Checking the efficiency of market (NIFTY 50) as proxy to Indian Market

The return(-1) ie the lag of the return is significant within 90% confidence interval.
h-Statistic: 0.006741810013810961

We therefore fail to reject the null hypothesis implying No autocorrelation exists: this supports that
the market is weak form efficient

You might also like