Terms : Hide Images
| 5671002872 | BSS/(k-1) | MSa | 0 | |
| 5671007627 | WSS/(n-k) | MSr(a) | 1 | |
| 5671013005 | MSa/MSr(a) df: k-1, n-k | F (calculating from sum of squares) | 2 | |
| 5671021009 | s1^2/s2^2 df: n1-1, n2-1 | F (different SS calc) | 3 | |
| 5671124570 | BSS/TSS; ESS/TSS | r^2 from ANOVA | 4 | |
| 5671080910 | (B-hat2 - B)/se of B-hat2 df: n-2 | t | 5 | |
| 5671168048 | r^2(n-2)/(1-r^2) df: 1, n-2 | F from r^2 | 6 | |
| 5671229664 | B | The term least squares, when used with regression analysis, refers to minimizing: a. the sum of the squares of the random error terms b. the sum of the squares of the residuals c. the sum of the deviations from the mean d. the square of the sum of the residuals e. the square of the sum of the random error terms | 7 | |
| 5671229688 | E | The correlation coefficient (r), or its square, calculated from a random sample of two variables: a. measures the covariation between the two variables in the sample b. can be used to determine if there are near multicollinearity problems between independent variables in a multivariate regression c. measures "goodness of fit" about the linear least-squares regression line for the observed values of the dependent variable in the sample d. can be used to estimate the correlation coefficient in the population e. all of the above | 8 | |
| 5671237554 | C | Which of the following is true concerning the F-test and t-test for multivariate regression equations: a. the t and F-tests yield identical results b. the t-test is used to test the statistical significance of the entire regression, whereas the F-test is used to test the statistical significance of the individual partial regression coefficients c. the F-test is used to test the statistical significance of the entire regression, whereas the t-test is used to test the statistical significance of the individual partial regression coefficients d. the t-test is a joint hypothesis test, whereas the F-test is not e. b and d | 9 | |
| 5671242819 | D | In a one-way analysis of variance (ANOVA), the within-groups sum of squares and the between-groups sum of squares respectively represent: a. the total variation and the unexplained variation b. the proportionate reduction of error and the marginal reduction of error c. the marginal reduction of error and the proportionate reduction of error d. the unexplained variation and the explained variation e. the explained variation and the unexplained variation | 10 | |
| 5671262944 | Two-variable regression model (bivariate) | a mathematical equation that defines the relationship between two variables; y = a + bx; it is used to predict y based on a selected x value; y is the dependent variable, and x is the independent variable | 11 | |
| 5671272222 | General linear regression model (multivariate) | the relationship in the form of a mathematical equation between several independent variables & a dependent variable; y-hat = a + b1x1 + b2x2 + b3x3 + ... + bkxk; it is used to estimate y given k independent variables (the x's) Additional assumption: linear independence of the variables | 12 | |
| 5671295284 | Residual term | the vertical distance between the regression line predicted value for Y and the observed value of Y in the sample (unexplained variation from your sample model) | 13 | |
| 5671304303 | Error term | the vertical distance between the regression line predicted value for Y and the observed value of Y in the population (unexplained variation from your population model - almost never used) | 14 | |
| 5671330258 | ANOVA | compares the differences between subsample means and the overall sample mean; each of these differences is a deviation from the overall mean, or the variance | 15 | |
| 5671337438 | Standard error of the regression | the average error made in predicting the dependent variable using the regression equation (the difference between y-hat and y) | 16 | |
| 5671358408 | BSS; explained sum of squares WSS; unexplained sum of squares | SAS output for regression Model = Error = | 17 | |
| 5671379221 | Type of wand As many as they had data on | NYC case study video What was the cause of the high water bills? Which variables did they include in the regression? | 18 | |
| 5671390397 | Resource dependency Traded & local cluster growth Cluster strength & diversity | Industry clusters & regional economic performance __________ had the strongest effect on economic development ___________ are statistically & practically significant ___________ are not significant | 19 | |
| 5671424900 | Mary Donegan et. al | Found that measures of creativity are generally NOT associated with differences in metropolitan economic performance; human capital & industry composition perform as well as or better than talent, tolerance, & technology in explaining metropolitan job & income growth & job stability | 20 | |
| 5671437917 | Near multicollinearity | Unbiased parameter estimates Inflated but still minimum variance t & F are good Can still use OLS estimators | 21 | |
| 5671454616 | Heteroscedasticity | Unbiased parameter estimates Inflated variance - no longer minimum variance t & F bad Cannot use OLS estimators | 22 | |
| 5671472716 | Serial Correlation | Unbiased parameter estimates Generally inflated variance - no longer minimum variance t & F are bad (biased upwards or downwards) Cannot use OLS estimators | 23 | |
| 5671483408 | Omission of a Relevant Variable | Biased parameter estimates Deflated but still minimum variance t & F bad Cannot use OLS estimators | 24 | |
| 5671494167 | Inclusion of an Irrelevant Variable | Unbiased parameter estimates Inflated but still minimum variance t & F good Can still use OLS estimators | 25 | |
| 5671521991 | Omission of a relevant variable | The worst violation of general linear model assumptions | 26 | |
| 5671565904 | Null hypotheses for a two-way ANOVA | Treatment: H0(A): μ1. = μ2. = ... = μa. Group: H0(B): μ.1 = μ.2 = ... = μ.b Interaction effect: H0(AB): μab = μ + (μa. - μ) + (μ.b - μ) for all combinations of a and b | 27 | |
| 5671604618 | ANOVA The calculated F statistic | Top part of regression is: Use sum of squares to get: | 28 | |
| 5671672032 | Logical 25% | R squared indicates a relationship when: Your model is: & r squared is around: | 29 | |
| 5671708861 | Calculated F > critical F OR p-value <= sig level Calculated F < critical F OR p-value > sig level | One-way ANOVA Reject null if: Do not reject null if: | 30 | |
| 5671784006 | Extending a one-way ANOVA to a two-way ANOVA | BSS = BcSS; main effects of Factor A Must find BrSS; main effects of Factor B Must find WgSS; unexplained influences & random error Must find ISS; interaction effects of Factors A & B | 31 | |
| 5671800646 | Extending a one-way ANOVA to a regression analysis | Sum of squares used to calculate F & r squared | 32 | |
| 5671837190 | The error term is random The sum of the random error will = 0 & the mean of the random error will = 0 | First assumption of the general (multivariate) linear regression model | 33 | |
| 5671866785 | Error terms cannot predict other error terms The error terms of the independent variables are independent | Second assumption of the general (multivariate) linear regression model | 34 | |
| 5671887038 | Error term is constant | Third assumption of the general (multivariate) linear regression model | 35 | |
| 5671891701 | Error term is independent of the independent variables The independent variable is not correlated with the error term | Fourth assumption of the general (multivariate) linear regression model | 36 | |
| 5671909013 | All of the independent variables are independent of each other | Fifth assumption of the general (multivariate) linear regression model | 37 | |
| 5671938694 | Independent variables are in the correct form & are the correct variables to be included | Sixth assumption of the general (multivariate) linear regression model | 38 | |
| 5671990900 | BSS + WSS | TSS = | 39 |
