Technical：问linear regression。非术语解释linear regression。
Linear regression: an approach to modelingthe relationship between a scalar dependent variable y and one or moreexplanatory variable x.
P-value: the probability ofobtaining a test statistic at least as extreme as the one that was actuallyobserved, assuming the null hypothesis is true. Aresearcher will often reject the null hypothesis when the P-value turns out tobe less than a certain significance level.
Type I Error: the incorrect rejectionof a true null hypothesis.
Type II Error: failure to reject afalse null hypothesis.
Mode: the most frequent orcommon score in the distribution.
Positively skewed: it has a tail extendingout to the right.
Negatively skewed: it has a tail extendingout to the left.
Standard deviation: it is an indicator of thevariability of a set of continuous scores around the mean.
Variance: The variance of a set of scores is simply thestandard deviation of that set of scores squared.
Statistical Inference: statisticalinference makes use of information from a sample to draw conclusions about thepopulation from which the sample was taken.
Central Limit TheoremThe Central LimitTheorem states that whenever a random sample of size n is taken from anydistribution with mean μ and variance file://localhost/Users/vivian/Library/Caches/TemporaryItems/msoclip/0/clip_image002.png, then the sample mean file://localhost/Users/vivian/Library/Caches/TemporaryItems/msoclip/0/clip_image004.png will be approximately normallydistributed with mean μ and variance file://localhost/Users/vivian/Library/Caches/TemporaryItems/msoclip/0/clip_image006.png/n. The larger the value of the samplesize n, the better the approximation to the normal. Quartile: values that divide a sample of data into fourgroups containing equal numbers of observation.Coefficient of Variation:t is the ratio of the sample standarddeviation to the sample mean:file://localhost/Users/vivian/Library/Caches/TemporaryItems/msoclip/0/clip_image008.pngQQ plot: it is used to see if a given set of datafollows some specified distribution. It should be approximately linear if thespecified distribution is the correct model.Power: the probability of rejecting the nullhypothesis when the null hypothesis is false.