# Interpreting integrity tests data on a regression analysis equation

I need someone to critically assess the relative merits/weaknesses of a economic modeling equation and the subsequent integrity tests performed on the 40 year time series data. Attached is a word document with screenshots of various EViews results/tests/diagrams, etc and an excel file with the associated data. Also attached is the EViews file I am using to analyze the data.

A brief comment on each screenshot, what it means, raise red flags (and yellow flags) of flaws in the data, etc.....and if heteroskedasticity, multicollinearity, autocorrelation is present how specifically to compensate for it.

Section 1 of the word doc uses the basic equation and Section 2 looks at the same equation with AR(1) AR(2) autoregression added.

© BrainMass Inc. brainmass.com October 9, 2019, 4:48 pm ad1c9bdddfhttps://brainmass.com/statistics/regression-analysis/interpreting-integrity-tests-data-regression-analysis-41424

#### Solution Preview

Please see the attachment for the response.

(DATA IS IN EXCEL SPREADSHEET)

1 - EQUATION

ESTIMATED EQUATION - STARTS DISPINCPC POPINC VCYRATE LOG(PRIMEANNUAL) C

SUBSTITUTED COEFFICIENTS PROVIDED BY EVIEWS ANALYSIS OF ESTIMATED EQUATION

STARTS = -910.1124445*LOG(PRIMEANNUAL) + 0.1067436277*DISPINCPC - 0.5215825002*POPINC - 624.0425476*VCYRATE + 5180.539466

The R-squared and the adjusted R-squared of the equation show that 60% and 55% of the variation can be, can be predicted using a quadratic function.

The f-statistic is large that is 13.5 and the p value is less than 0.05 means that your model has some value and green flag is on.

The t statistic in case of two variables is 6.4 and 10.1 that means that the predicted values are not large enough. On the other hand the large negative values of -5.6, -5.7 and -6.1 indicate that the hypothesized value of POPINC, VCYRATE AND PRIMEANNUAL are too large. This is clearly a yellow flag. As you can see that the standard error in case of the last three variables is really large indicating that the sample size is small. This can be compensated by reducing the number of variables. Remember, Heteroskedasticity often occurs when there is a large difference between the size of observations.

1. [1] cites a cross sectional example: Comparing states with widely differing populations, such as Rhode Island and California.

Imagine you are watching a rocket take off nearby and measuring the distance it has travelled once each second. In the first couple of seconds your measurements may be accurate to the nearest centimeter, say. However, 5 minutes later as the rocket recedes into space, the accuracy of your measurements may only be good to 100m, because of the increased distance, atmospheric distortion and a variety of other factors. The data you collect would exhibit heteroskedasticity.

* Durbin-Watson stat is significant so we address this later on using AR(1) and AR(2)

The R-squared and the adjusted R-squared of the equation show that 66% and 61% of the variation can be, can be predicted using a quadratic function.

The f-statistic is large that is 13.7 more than the critical level of 2.85 and the p value is less than 0.05 means that your model has some value and green flag is on.

The t statistic in case of four variables is 1.6,1.7,1.7 and 2.5 that means that the predicted values are not large enough. On the other hand the large negative values of -1.6, and -1.5 indicate that the hypothesized value of the variables are too large. This is clearly a yellow flag. In case of variable 1,4 and 5 the ...

#### Solution Summary

The expert interprets integrity test on a regression analysis. The solution answers the question(s) below.