Statistics and Quantitative Analysis U4320 Lecture 13: Explaining Variation Prof. Sharyn O’Halloran
I. Explaining Variation: R2 • A. Breaking Down the Distances • Let's go back to the basics of regression analysis. • How well does the predicted line explain the variation in the independent variable money spent?
I. Explaining Variation: R2 • Total Variation
I. Explaining Variation: R2 • Total Deviation Total = Explained + Unexplained Deviation Deviation Deviation • The total distance from any point to is the sum of the distance from Y to the regression line plus the distance from the regression line to .
I. Explaining Variation: R2 • B. Sums of Squares • We can sum this equation across all the Y's and square both sides to get:
I. Explaining Variation: R2 • 1. Total Sum of Squares (SST). • The term on the left-hand side of this equation is the sum of the squared distances from all points to . We call this the total variation in the Y's, or the Total Sum of Squares (SST). • 2. Regression Sum of Squares • The first term on the right hand side is the sum of the squared distances from the regression line to . We call it the Regression Sum of Squares, or SSR.
I. Explaining Variation: R2 • 3. Error Sum of Squares • Finally, the last term is the sum of the squared distances from the points to the regression line. Remember, this is the quantity that least squares minimizes. We call it the Error Sum of Squares, or SSE. • We can rewrite the previous equation as: SST = SSR + SSE.
I. Explaining Variation: R2 • C. Definition of R2 • We can use these new terms to determine how much variation is explained by the regression line. • If the points are perfectly linear, then the Error Sum of Squares is 0:
I. Explaining Variation: R2 • Here, SSR = SST. The variance in the Y's is completely explained by the regression line.
I. Explaining Variation: R2 • On the other hand, if there is no relation between X and Y: • Now SSR is 0 and SSE=SST. The regression line explains none of the variance in Y.
I. Explaining Variation: R2 • 3. Formula • So we can construct a useful statistic. • Take the ratio of the Regression Sum of Squares to the Total Sum of Squares: • We call this statistic R2 • It represents the percent of the variation in Y explained by the regression
I. Explaining Variation: R2 • R2 is always between 0 and 1. • For a perfectly straight line it's 1, which is perfect correlation. • For data with little relation, it's near 0. • R2 measures the explanatory power of your model. • The more of the variance in Y you can explain, the more powerful your model.
I. Explaining Variation: R2 • D. Example • I wanted to investigate why people have confidence in what they see on TV. • 1. Dependent variable • TRUSTTV = 1 if has a lot of confidence = 2 if somewhat confidence = 3 if the individual has no confidence. • 2. Independent variables • TUBETIME = number of Hours of TV watched a week • SKOOL = years of education. • LIKEJPAN = feelings towards Japan. • YELOWSTN = attitudes whether the US should spend more on national parks. • MYSIGN = the respondents astrological sign.
I. Explaining Variation: R2 • 3. Calculating R2 • a) Correlation matrix
I. Explaining Variation: R2 • b) The first model is: TRUSTTV = 2.34 - 0.0539 (TUBETIME)
I. Explaining Variation: R2 • How do we calculate the Total Sum of Squares? SST = SSR + SSE SST = 5.90 + 183.61 = 189.51 • Now we can calculate R2:
I. Explaining Variation: R2 • c) Each of the 4 different models has an associated R2. Eq # 2: R2 = 0.035 Eq # 3: R2 = 0.039 Eq # 4 = R2 = .04
I. Explaining Variation: R2 • F. Using R2 in Practice • 1. Useful Tool • 2.Measure of Unexplained Variance • 3. Not a Statistical Test • 4. Don't Obsess about R2 • 5. You can always improve R2 by adding variables
I. Explaining Variation: R2 • G. Example • You'll notice that R2 increase every time. • No matter what variables you add you can always increase your R2.
II. Adjusted R2 • II. Adjusted R2 • A. Definition of Adjusted R2 • So we'd like a measure like R2, but one that takes into account the fact that adding extra variables always increases your explanatory power. • The statistic we use for this is call the Adjusted R2, and its formula is: • So the Adjusted R2 can actually fall if the variable you add doesn't explain much of the variance.
II. Adjusted R2 • B. Back to the Example • 1. Adjusted R2 • You can see that the adjusted R2 rises from equation 1 to equation 2, and from equation 2 to equation 3. • But then it falls from equation 3 to 4, when we add in the variables for national parks and the zodiac.
II. Adjusted R2 • 2. Calculating Adjusted R2 • Example: Equation 2
II. Adjusted R2 • We calculate:
II. Adjusted R2 • C. Stepwise Regression • One strategy for model building is to add variables only if they increase your adjusted R2. • This technique is called stepwise regression. • However, I don't want to emphasize this approach to strongly. Just as people can fixate on R2 they can fixate on adjusted R2. ****IMPORTANT**** If you have a theory that suggests that certain variables are important for your analysis then include them whether or not they increase the adjusted R2. Negative findings can be important!
III. F Tests • III. F Tests • A. When to use an F-Test? • Say you add a number of variables into a regression model and you want to see if, as a group, they are significant in explaining variation in your dependent variable Y. • The F-test tells you whether a group of variables, or even an entire model, is jointly significant. • This is in contrast to a t-test, which tells whether an individual coefficient is significantly different from zero.
III. F Tests • B. Equations • To be precise, say our original equation is: EQ 1: Y = b0 + b1X1 + b2X2, and we add two more variables, so the new equation is: EQ 2: Y = b0 + b1X1 + b2X2 + b3X3 + b4X4. • We want to test the hypothesis that H0:b3 = b4 = 0. That is, we want to test the joint hypothesis that X3 and X4 together are not significant factors in determining Y.
III. F Tests • C. Using Adjusted R2 First • There's an easy way to tell if these two variables are not significant. • First, run the regression without X3 and X4 in it, then run the regression with X3 and X4. • Now look at the adjusted R2's for the two regressions. If the adjusted R2 went down, then X3 and X4 are not jointly significant. • So the adjusted R2 can serve as a quick test for insignificance.
III. F Tests • D. Calculating an F-Test If the adjusted R2 goes up, then you need to do a more complicated test, F-Test. • 1. Ratio • Let regression 1 be the model without X3 and X4, and let regression 2 include X3 and X4. • The basic idea of the F statistic, then, is to compute the ratio:
III. F Tests • 2. Correction • We have to correct for the number of independent we add. • So the complete statistic is: • Remember that k is the total number of independent variables, including the ones that you are testing and the constant.
III. F Tests • 2. Correction (cont.) • This equation defines an F statistic with m and n-k degrees of freedom. • We write it like this: • To get critical values for the F statistic, we use a set of tables, just like for the normal and t-statistics.
III. F Tests • E. Example • 1. Adding Extra Variables: Are a group of variables jointly significant? • Are the variables YELOWSTN and MYSIGN jointly significant?
III. F Tests • 1. Adding Extra Variables (cont.) • a) State the null hypothesis • b) Calculate the F-statistic • Our formula for the F statistic is:
III. F Tests • What is SSE1 --the sum of squared errors in the first regression? • What is SSE2, the sum of squared errors in the second regression? m = 2 N = 470 k = 6 • The formula is:
III. F Tests • c) Reject or fail to reject the null hypothesis? • The critical value at the 5 % level • from the table, is 3.00. • Is the F-statistic > ? • If yes, then we reject the null hypothesis that the variables are not significantly different from zero; otherwise we fail to reject. • .319 < 3.00, so we can reject the null hypothesis.
III. F Tests • 2. Testing All Variables: Is the Model Significant? • Equation 2
III. F Tests • a) Hypothesis: • Again, we start with our formula:
III. F Tests • b) Calculate F-statistic • SSE2 = 182.78 • SSE1 is the sum of squared errors when there are no explanatory variables at all. If there are no explanatory variables, then SSR must be 0. In this case, SSE=SST. • So we can substitute SST for SSE1 in our formula. SST = SSR + SSE = 6.738 + 182.78 = 189.54 • This is the number reported in your printout under the F statistic.
III. F Tests • c) Reject or fail to reject the null hypothesis? • The critical value at the 5% level, from your table, is 3.00. • So this time we can reject the null hypothesis that b1 = b2 = 0.