Question

In: Economics

Why is not possible to employ OLS to estimate the parameters of a non stationary stochastic...

Why is not possible to employ OLS to estimate the parameters of a non stationary stochastic process?

Solutions

Expert Solution

Answer. Time series that diverge away from their mean over time are said to be non-stationary. Therefore, the classical estimation of variables with this relationship most times gives misleading inferences or spurious regression.
A non-stationary time series is a stochastic process with unit roots or structural breaks. However, unit roots are major sources of non-stationarity. The presence of a unit root implies that a time series under consideration is non-stationary while the absence of it entails that a time series is stationary. This depicts that unit root is one of the sources of non-stationarity. A non-stationary stochastic process could be Trend Stationary (deterministic) Process (TSP) or Difference Stationary Process (DSP).
It should be made clear that if a time series is TSP, but treated as DSP, this is called over-differencing. On the other hand, if a time series is DSP, but treated as TSP; this is referred to as under-differencing. The implications of these types of specification error can be serious, depending on how the serial correlation properties of the resulting error terms are handled. However, it has been observed that most time series are DSP rather than TSP. Therefore, when such non-stationary time series (DSP) are used in estimation of an econometric model, the Ordinary Least Square (OLS) traditional diagnostic statistics for evaluation of the validity of the model estimates such as, coefficient of determination (R2), Fisher’s Ratio(F Statistic), Durbin-Watson(DW-Stat), t-statistic etc. become highly misleading and unreliable in terms of forecast and policy. In such series, the mean, variance, covariance and autocorrelation functions change overtime and affect the long run development of the series. The presence of unit root in these series leads to the violation of assumptions of constant means and variances of OLS.


Related Solutions

OLS Mechanics 2. Describe the difference in interpretation, if there is any, between the population parameters...
OLS Mechanics 2. Describe the difference in interpretation, if there is any, between the population parameters (β0 and β1) and estimated coefficients (βˆ 0 and βˆ 1). 3. Describe what information βˆ 1 and ρ can reveal. What information do they both share? And what distinct information do they provide?
What are the problems of analyzing non-stationary data?
What are the problems of analyzing non-stationary data?  
Explain the difference between a non-stationary variable and a stationary variable. Give examples for both of...
Explain the difference between a non-stationary variable and a stationary variable. Give examples for both of them. then Explain the difference between a stationary variable and white noise variable. Give an example of a white noise variable. Why is a white noise variable also a stationary variable? Explain
Can we transform a non-stationary process into a 1st order weakly stationary or 2nd order weakly...
Can we transform a non-stationary process into a 1st order weakly stationary or 2nd order weakly stationary/covariance-stationary process?
If you estimate the beta OLS for stock i is 1.5, the standard error is 0.3....
If you estimate the beta OLS for stock i is 1.5, the standard error is 0.3. To test the two null hypothesis:β=3.5,β=-1,can you reject these two null hypotheses at 10% significance level with the tcritical = 2.96? A. β=3.5 cannot be rejected,β=-1 can be rejected B. β=3.5 can be rejected,β=-1 cannot be rejected C. β=3.5 can be rejected,β=-1 can be rejected D. β=3.5 cannot be rejected,β=-1 cannot be rejected
If you estimate the beta OLS for stock i is 1.5, the standard error is 0.3....
If you estimate the beta OLS for stock i is 1.5, the standard error is 0.3. To test the two null hypothesis:β=3.5,β=-1,can you reject these two null hypotheses at 10% significance level with the tcritical = 2.96? A. β=3.5 cannot be rejected,β=-1 can be rejected B. β=3.5 can be rejected,β=-1 cannot be rejected C. β=3.5 can be rejected,β=-1 can be rejected D. β=3.5 cannot be rejected,β=-1 cannot be rejected
We use sample data to estimate population parameters and point estimate and interval estimate are two...
We use sample data to estimate population parameters and point estimate and interval estimate are two typical tools of such estimation. Please discuss them answering the following questions. 1) Briefly discuss about how point estimate and confidence interval are different to each other. 2) Discuss about way(s) to secure unbiased and efficient point estimators. 3) Discuss about the effects of confidence level (or, Z-score) and sample size on the width of confidence interval. Is it good or bad to have...
why does endogeneity lead to biased OLS estimates?
why does endogeneity lead to biased OLS estimates?
Suppose you estimate the following regression model using OLS: Yi = β0 + β1Xi + β2Xi2...
Suppose you estimate the following regression model using OLS: Yi = β0 + β1Xi + β2Xi2 + β3Xi3+ ui. You estimate that the p-value of the F-test that β2= β3 =0 is 0.01. This implies: options: You can reject the null hypothesis that the regression function is linear. You cannot reject the null hypothesis that the regression function is either quadratic or cubic. The alternate hypothesis is that the regression function is either quadratic or cubic. Both (a) and (c).
What assumption can you use the binary variable specification and OLS to estimate the combined entity...
What assumption can you use the binary variable specification and OLS to estimate the combined entity and time fixed effects regression model? What is your suggested solution?
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT