COMPSCI C8 Lecture Notes - Lecture 25: Standard Deviation, Histogram

• School: Berkeley
• Department: Computer Science
• Course: COMPSCI C8
• Professor: John Denero

Document Summary

Standard deviation (sd) measures roughly how far the data from their average. Sd = root mean square of deviations from average. D has the sa(cid:373)e u(cid:374)its as the data; he(cid:374)(cid:272)e ok to say (cid:862)a(cid:448)e(cid:396)age plus o(cid:396) (cid:373)i(cid:374)us a fe(cid:449) ds(cid:863) No matter what the shape of the distribution, the proportion of values in the range (cid:862)a(cid:448)e(cid:396)age +- z ds(cid:863) is: at least 1 1/z^2. How many sds above average? z = (value mean)/sd: negative z: value below average, positive z: value above average, z = 0: value equal to average. When values are in standard units: average = 0, sd = 1. Most values of z between -5 and 5. It"s (cid:374)ot easy to esti(cid:373)ate d (cid:271)y looki(cid:374)g at histog(cid:396)a(cid:373) If histogram has bell shape, then you can. If a histogram is bell-shaped, then: the average is at the center, the sd is the distance between the average and the points of inflection on either side.