KEMBAR78
Steps Ofvvector Estimating Error Correction Model | PDF | Regression Analysis | Statistical Analysis
Scribd
Upload
146 views4 pages

Steps Ofvvector Estimating Error Correction Model

This document outlines the steps for vector error correction model estimation including: 1) specifying the model and preparing data for time series analysis, 2) testing for stationarity and determining the optimal lag length, 3) performing cointegration tests to determine whether to use a VAR or VECM model, and 4) diagnostic testing and corrections for issues like autocorrelation, heteroskedasticity, and model specification.

Uploaded by

Namdev Upadhyay
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
146 views4 pages

Steps Ofvvector Estimating Error Correction Model

This document outlines the steps for vector error correction model estimation including: 1) specifying the model and preparing data for time series analysis, 2) testing for stationarity and determining the optimal lag length, 3) performing cointegration tests to determine whether to use a VAR or VECM model, and 4) diagnostic testing and corrections for issues like autocorrelation, heteroskedasticity, and model specification.

Uploaded by

Namdev Upadhyay
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Steps ofvVector Estimating Error correction model

1. Specify model: Ovtest


2. prepare stata for for time series analysis
3. stationary test
4. determine optimum lag length (P) for model
5. perform johnsen cointegration test
6. with no cointegration: VAR model
7. with co-integration; VECM model
8. perform diagnostic
A. Augmented Dikkey Fuller test (ADF): Finding I(0) or I(1)
1. Set time series data: tsset year, yearly
2. Fill missing data for Time Series analysis: ipolate x time, gen(xi) epolate
3. Check unit root using Augmented Dikker Fuller test (ADF): at integrative of order zero
ADF unit root test using constant
dfuller x
ADF test for constant and trend:
dfuller x, trend
if a variable has unit root, we take difference as follows:
dfuller d.x
ADF test after differencing for constant and trend:
dfuller d.x, trend

B. Checking assumption of linear ordinary least regression


1. If you want to have summary statistics: sum y x1 x2 x3 x4
2. How to run correlation matrix for your data: correlate y x1 x2 x3 x4
3. Command for running regression model: regress y x1 x2 x3 x4
4. If you want to check normality after running regression model, run two commands consecutively:
 predict yhat, r
 sktest yhat
5. Testing of Auto correlation

Visual method

 regress y x1 x2 x3 x4
 predict yhat, r
 tsline yhat

Clusttering of value

dwstat
Correction

1. First difference transformation


 reg d.y d.x1 dx2
 predict dhat, r
 tsline dhat
2. Generalized list square approach
 prais y x, corc

if about this does not eliminate the problem of auto correlation

3. Newey-West Robust standard Error


 neway y x, lag(1)
4. Heteroscedasticity

estat hettest

Note: Prob value below 10% means there is heteroskedasticity problem in the model:

Correction Measure

 Log transformation of variable


 Weighted list Method: reg y x (aw=1/one of the variable) trail and error process
 robus option

reg y x, robus

estato robust

 re-estimate the model with robus option


 estato ols
 estatout
5. Model Specification: estat ovtest
C. Determine Optimum lag
Command for selecting optimum lags for your model is given below:

varsoc y x1 x2 x3 x4, maxlag(4)

the asterisk (*) indicates the appropriate lag selected

D. VECM model

Command for testing co-integration:


vecrank y x1 x2 x3 x4, trend(constant)
When co-integration is established, run VECM otherwise unrestricted VAR model is
appropriate.
Assuming variables are co-integrated, we run VECM using the following command:
vec y x1 x2 x3 x4, trend(contant)
The long-run causality must be negative, significant and in between 0 to 1, representing
error correction term or speed of adjustment
Correction for auto-correlation

 First difference transformation


 Generalized difference

Drop missing variable

Drop Year=.

You might also like