DESCRIPTIVE STATISTICS

Descriptive statistics is the fundamental requirement for any data analysis. Exploratory Data Analysis (EDA) is performed in almost all data science projects to understand the nature of the data. In this article, I would like to discuss two underlying concepts involved in descriptive statistics

1. Estimate of Location: This is the technique performed to understand where most of the data is clustered.
2. Estimate of variability (Deviation): This is a way to identify, how dispersed is the value, or how cluttered the value from the median or mean.

Now let's take a deeper look at the individual concepts of Estimate of Location and Estimate of variability

ESTIMATE OF LOCATION:

1. Mean: is the average value of the dataset. it is calculated by adding or summing the value field and dividing the total sum by the number of rows.

For Instance: 2+4+4+2= 12

We have 4 records and the total sum is 12, hence the mean value would be 3

1. Median: This is a middle value in a dataset. This is the more robust value when compared to the mean because it takes in the consideration of outliers in the dataset.
2. Mode: This is how many times a particular value has been repeated in the dataset.

For Instance: 1, 2, 2, 4, 5, 6

2 is the mode in the above example because 2 is repeated more times.

ESTIMATE OF VARIABILITY:

1. Range: This is the difference between the maximum value and the minimum value.

For instance: Max value = 10 and the Min value = 2 in a given variable, 8 is the range

1. Variance: This is a measure of how the data differs from the mean value.
2. Deviation Score: This is the difference between the number of observed values and the estimate of location (Mean or Median)
3. Standard Deviation: This is simply the square root of the variance

For Instance: if the variance is 34.8 then the Standard deviation is 5.89 (Square root of 34.8)

1. Percentile: This is the place of the value in a given dataset. (This is also a way of ranking observations)

For instance: Mr. X scored 75th percentile, which means 75% of the people scored less than Mr. X

1. Interquartile Range: The difference between 75th and 25th percentile. This is also nothing but middle 50th percentile or 50th Percentage