<Student’s name>
<Professor’s name>
Introduction
In this paper we are interesting in determining a probability distribution for the length of a call to this call center based on the historical data. To do this, we consider the sample descriptive statistics, including measures of central tendency and measures of variability. We will construct charts and histograms to understand the probability distribution and run some statistical tests to understand the distribution law.
Descriptive Statistics
Descriptive Statistics: length
Variable N N* Mean SE Mean StDev Variance CoefVar Minimum Q1
length 80 0 196,6 38,2 342,0 116978,7 173,99 1,0 54,3
N for
Variable Median Q3 Maximum Range IQR Mode Mode
length 103,5 200,5 2631,0 2630,0 146,3 9 3
Charts and Graphs
Let’s test the sample on normality using Anderson-Darling test. We obtain the following result on a plot:
Since p-value is lesser than 0.005, the data seems to be non-normal.
According to our histograms and charts we may do a conclusion, that there are couple of residuals in the data (see boxplot). Let’s exclude those points from a data sample and test for normality once again. Outliers are all points from 438 and above.
The new data set has the following visual representation:
But the normality test gives us the same result:
So, the data is not normal. The probability distribution of the data set is explained by descriptive statistics and histograms and charts.