2.5 Guessing The Distribution of Dataset 2
2.5 Guessing The Distribution of Dataset 2
Uniform Distribution
• Peaks on either side, but it's not symmetric in nature, in the sense there is no stepwise decrease
on either side.
• So, it's a non-Gaussian, somewhat symmetric.
• Between the minimum and 25%, 25% to 50%, 50% to 75%, and 75% to maximum, the intervals
are around 2.
mi n + max 50 + 60 110
• = = = 55 = mean . We can assume that this comes from uniform which is
2 2 2
between 50 and 60 .
• From these numbers, we can conclude that it's a uniform distribution. Because between the
quantiles, the split is almost the same, and the standard deviation is very small.
• Kurtosis is negative.
• Generally for normal distributions the Kurtosis would be around 3. So it cannot be a normal.
• Skewness is very low. So, it's not skewed to either side. It is centred.
From both plots, we can see that the observed points are very close to the red line.
So, it's a uniform distribution.
ALTERNATE HYPOTHESIS: The given data does not follow Uniform distribution
• p-value > α
We accept the null.
df = k − p − 1 = 50 − 0 − 1 = 49
k = 50 : number of observations
Case examples • Throwing a die. The probability of getting each side is 16 . It's a uniform distribution.
• Fuel Efficiency. If you put one litre petrol/diesel, every time you would get a different mileage. It
could be 50, 52, 47. In case you're a rash driver, it could be low, say 40, 42.