STATA

10
STATA Introductory manual

description

STATA. Introductory manual. QUIZ. What are the main OLS assumptions? On average right Linear Predicting variables and error term uncorrelated No serial correlation in error term Homoscedasticity + Normality of error term. QUIZ. Do we know the error term? Do we know the coefficients? - PowerPoint PPT Presentation

Transcript of STATA

Page 1: STATA

STATA

Introductory manual

Page 2: STATA

QUIZ

• What are the main OLS assumptions?1. On average right2. Linear3. Predicting variables and error term uncorrelated4. No serial correlation in error term5. Homoscedasticity+ Normality of error term

Page 3: STATA

QUIZ

• Do we know the error term?• Do we know the coefficients?• How can we know whether all the

assumptions are fulfilled1. On average right => ???2. Linearity => ???3. X and ε uncorrelated => ???4. ε serially uncorrelated => ???5. ε homoscedastic => ???

Page 4: STATA

STATA as calculator

• Multiplication and other simple operations– display (7*(6+5)/4)^(-0.3)

• Statistical calculations– display norm(1.96)

• Generating variables– gen t=year-1967

• Etc...

Page 5: STATA

STATA for matrices

• Writing down matrix– matrix X=(1,4\1,-2\1,3\1,-5)

• Reading a matrix (you have to know its name)– matrix list X

• Learning what matrices there are– matrix dir

• Multiplications – matrix XX=X’*X– matrix IXX=inv(XX)

• Etc...

Page 6: STATA

Getting data to STATA

• How to get data to STATA?– From Excel [Copy&Paste, remember „commas”]– Manually [if one has the time ... ]– From other sources [import function]

• How to know what it is?– Describe [read the texts]– summarize [statistical properties]

• Learning more about your data set– correlate x1 y [just to see how it works]– histogram x1 [graphing is easier from the menu]– scatter x1 y [as above ]

Page 7: STATA

Regression with STATA

• How to do regression?– regress y x– regress y x, noconsand that’s it

Page 8: STATA

Diagnostics with STATA

• Normality of the residual– predict e, residual [directly after regress]– sktest e [Jarque-Bery test]

• RESET test– ovtest, rhs

Page 9: STATA

Diagnostics with STATA

• Heteroscedasticity– hettest, rhs [Breush-Pagan test]– imtest, white [White test]

• Autocorrelation– tsset t – dwstat [Durbin-Watson test]– bgodfrey, lags(1 2 3) [Breush-Godfrey test]

Page 10: STATA

Diagnostics with STATA

• Structural stability [Chow]– gen d=0– gen dx1=0– gen dx2=0– gen dx3=0– replace d=1 if t>50– replace dx1=x1 if t>50– replace dx2=x2 if t>50– replace dx3=x3 if t>50– reg y x1 x2 x3 d dx1 dx2 dx3– test (d=0) (dx1=0) (dx2=0) (dx3=0)