Stats II 3

11 flashcards · Other · by Daniel F.

1. Parts of Multivariate model

2. Parameter estimation

   Input y i: observations of the dependent variable Y (dependent or explained, or endogenous variable, regressand) x ki: observations of the independent variables of Xk (independent or explanatory, or exogenous variable, regressor, predictor) Output b0: estimated intercept bk: estimated slope of the independent variables of Xk

3. OLS method

   Ordinary least squares method minimizes unexplained variation

4. Interpretation of the parameters

   One X1 is expected to increase the Y by b1 ceteris paribus.

5. What conclusions can we make based on the confidence intervals

   With 95% confidence, the population value in within the interval constructed around the significant sample value

6. Test the regression

   Confidence interval on the population parameters Hypothesis testing on the individual parameters Overall significance

7. coefficient of determination R2

   the percentage of variability of the dependent variable explained by the variability of the independent variables

8. overall significance of regression

   H0: βj = 0 for all j = 1, ..., k H1: Not all βj = 0 H0: R2 = 0 H1: R2 ≠ 0

9. Beta vs B

   Beta - Standardize regression variables Which independent variable has the highest impact on the dependent variable normal standardization B - Raw coefficient

10. Outlier

    Observation that lies outside the overall pattern of a distribution Causes Procedural error Extraordinary event or variable realization Unique combination of the independent variables

11. Outliers on normally distributed data

    distribution of data 1sigma: 68% 2sigma: 95% 3sigma: 99% further away value is more likely outlier calculate: compare z-score with threshold small samples 2.5 larger samples 3.5