Download Cheat Sheet for Statistics Formulas and Calculator Instructions and more Lecture notes Technology in PDF only on Docsity! Page 1 of 2 I will provide these cheat sheets on the day of the final exam. Do not bring this copy to the final. Also, it is your responsibility to verify that the formulas are correct and notify me of any necessary corrections. Formulas Data Distribution ?ฬ
?๐ฅ = โ๐ฅ๐ฅ ๐๐ sd = ๏ฟฝโ(๐ฅ๐ฅโ?ฬ
?๐ฅ)2 ๐๐โ1 Probability Distribution ๐๐ = โ๐ฅ๐ฅ โ ๐๐(๐ฅ๐ฅ) ๐๐ = ๏ฟฝโ(๐ฅ๐ฅ โ ๐๐)2 โ ๐๐(๐ฅ๐ฅ) ๐๐ = ๐ฅ๐ฅโ๐๐ ๐๐ , ๐ฅ๐ฅ = ๐๐ + ๐ง๐ง โ ๐๐ Empirical Rule = 68-95-99.7 Rule See also the common probability rules below. Regression Predicted ๐ฆ๐ฆ = ๐๐+ ๐๐ โ ๐ฅ๐ฅ ๐๐ = ๐๐ โ ๐ ๐ ๐ฆ๐ฆ ๐ ๐ ๐ฅ๐ฅ , ๐๐ = ๐ฆ๐ฆ๏ฟฝ โ ๐๐ โ ?ฬ
?๐ฅ Residual (predicted error) = ๐ฆ๐ฆ๐๐๐๐๐ ๐ ๐๐๐๐๐๐๐๐๐๐ โ ๐ฆ๐ฆ๐๐๐๐๐๐๐๐๐๐๐๐๐๐๐๐๐๐ S๐๐ = ๏ฟฝ SSE ๐๐โ2 Distribution of Sample Proportions Mean = P, SE = ๏ฟฝ๐๐โ(1โ๐๐) ๐๐ , ๐ง๐ง = ๐๐๏ฟฝโ๐๐ ๐๐๐๐ Normal if ๐๐๐๐ โฅ 10 and (1 โ ๐๐) โ ๐๐ โฅ 10 CI = ๐๐๏ฟฝ ยฑ ๐๐๐๐ โ SE Distribution of Sample Means Mean = ๐๐, ๐ง๐ง = ?ฬ
?๐ฅโ๐๐ ๐๐๐๐ ๐๐ known & population variable is normally distributed OR n > 30: Use Z-test SE = ๐๐ โ๐๐ , CI = ?ฬ
?๐ฅ ยฑ ๐๐๐๐ โ SE ๐๐ not known & population variable is normally distributed OR n > 30: Use T-test One sample: SE = ๐ ๐ โ๐๐ , ๐๐๐๐ = ๐๐ โ 1 , CI = ?ฬ
?๐ฅ ยฑ ๐๐๐๐ โ ๐๐๐๐ Two independent samples: SE = ๏ฟฝ๐ ๐ 12 ๐๐1 + ๐ ๐ 22 ๐๐2 , ๐๐๐๐ (use technology) CI = (?ฬ
?๐ฅ1 โ ?ฬ
?๐ฅ2) ยฑ ๐๐๐๐ โ ๐๐๐๐ Chi-Square Tests All chi-square curves are skewed right; mean = ๐๐๐๐. Test statistic for all Chi-Square tests: ๐๐2 = (๐๐๐๐๐ ๐ ๐๐๐๐๏ฟฝ๐๐๐๐โ๐๐๐ฅ๐ฅ๐๐๐๐๐๐๐๐๐๐๐๐)2 ๐๐๐ฅ๐ฅ๐๐๐๐๐๐๐๐๐๐๐๐ One-way table: ๐๐๐๐ = (๐๐ โ 1) where r = number of categories. Two-way table: ๐๐๐๐ = (๐๐ โ 1)(๐๐ โ 1) where r = number of categories for one variable and c = number of categories for the other variable. Hypothesis Testing ๐ป๐ป0 : โค,โฅ, = (then change to =) ๐ป๐ป๐๐: <, >,โ (< or > is a one-tailed test; โ is a two-tailed test) Compare P-value to ๐ผ๐ผ One-tailed test: ๐ผ๐ผ = significance level Two-tailed test: ๐ผ๐ผ = significance level 2 Z-test: SE = ๐๐ โ๐๐ , ๐ง๐ง = ?ฬ
?๐ฅโ๐๐0 ๐๐๐๐ 1-sample T-test: SE = ๐ ๐ โ๐๐ , ๐๐ = ?ฬ
?๐ฅโ๐๐0 ๐๐๐๐ 2-sample T-test: SE = ๏ฟฝ๐ ๐ 1 2 ๐๐1 + ๐ ๐ 2 2 ๐๐2 ๐๐ = (?ฬ
?๐ฅ1โ?ฬ
?๐ฅ2)โ(๐๐1โ๐๐2) ๐๐๐๐ Paired T-test: SE = ๐ ๐ โ๐๐ , ๐๐ = ?ฬ
?๐ฅโ0 ๐๐๐๐ Probability Rules a) For any event A, 0 โค ๐๐(๐ด๐ด) โค 1 . b) If S is a sample space, then P(S) = 1. c) The sum of the probabilities of all possible disjoint events in a sample space is 1. d) If A and B are disjoint events (no outcomes in common), then P(A or B) = P(A) + P(B). e) If A and B are NOT disjoint events (share at least one outcome), then P(A or B) = P(A) + P(B) โ P( A and B). f) For any event A, P(not A) = 1 โ P(A). g) If P(B|A) โ P(B) , then A and B are independent events. h) P(A and B) = P(A)โP(B|A). i) When A and B are independent events then P(A and B) = P(A)โP(B)