Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

Normal Size Small Size show me how

# Regression

### Misc

Question | Answer |
---|---|

Three major factors determine how much beta hat varies across samples: | 1)Goodness of fit of the equation. 2)Range of observations on the independent variable. 3)Sample size. |

Type I Error | Rejecting the null when it’s true. The level of significance is the amount of Type I error implied by a test. |

Type II Error | “Accepting” the null when it’s false. |

4 things to know about Logs | 1) The log is the exponent of e. 2) A one unit change in the logarithm of a number is a one percent change in the number itself. 3) A one unit change is in the second decimal place in the log. 4) Adding logarithms of numbers is the same as multiplying |

What is the difference between Bias and Variance? | Bias messes with regressor coefficients by shifting your distribution. Variance messes with regressor coefficients by spreading distribution (increasing distance between X and Xbar. |

SE depends on what 4 factors? | 1) Goodness of fit of equation (RSS). 2) Size of equation. 3) Who widespread X’s are from their mean. 4) Correlation between two variables. |

What happens to SE as sample size increases? | SE will decrease. |

How do you interpret logs in table notes? | When the log is on the left, move the decimal to the right (2 places). When the log is on the right, move the decimal to the left (2 places). Changing the functional form of dependent variable makes Adjusted R2s incomparable. |

How do Intercept Dummies affect the model? | Cause a shift in the regression line. |

How do Slope Dummies affect the model? | Cause a tilt in the regression line. |

Created by:
kristel387