Kurtosis is used as a test for normality (covered later) due to the ability to detect the proportion of data that is an inlier and outlier. For a normal distribution (covered below), kurtosis takes the value 3 and so we can compare to this. We often define the excess kurtosis as: Excess Kurtosis = Kurtosis - 32. a. Skew and kurtosis are higher order descriptions of the data. Normal curves have known values for theser and so if a sample deviates signicantly, we can use them to detect this deviation. For the data in JBData.xlsx g, which consists of 2500 data points, calculate the skew and excess kurtosis, and then combine the values into the following formula for J B: J3 = % ($1:ve + E(Excess Kurtosisf) where for us, N = 2500. Large values (in this case, greater than approximately 6} for the number JB indicate a deviation from a normal distribution. b. Stretch: Visualise the data through a frequency plot {histogram}. Write a sentence or two regarding the plot and the number JB on whether you think the data follows a normal distribution. File Home Insert Page Layout Formulas Data Review View Help Bl - 5 >1 V/ 15' =SKEW(A1:A2500] A B | c | D | E \\ F | G | H | I \\ J | K | L | 0497113 051498048; ' ' ' ' ' ' ' ' 0.418523 0.255172. 0.121332E 0.175762- 0.433505. "I'ln'lnD q'm'm'h'w'w'd L