You are on page 1of 3

BI Calculation and statistic analytics Report

2
( y− y )
MSE=∑
n
RMSE=√ MSE

Training set:
(42,173) (49,198)
P (42) =190 P (49) =221
(190-173)= (17) ^2=289 (221-198) = (23) ^2=529
(37,149)
P (37) =168
(168-149)= (19) ^2 =361
(46,185)
P (46) =208
(208-185)= (23) ^2= 529
(30,123)
P (30) =137
(137-123)= (14) ^2=196
(50,201)
P (50) =225
(225-201)= (24) ^2=576
(43,174)
P (43) =194
(194-174)= (20) ^2=400
(43,175)
(194-175)^2=361
(46,188)
P (64) =208
(208-188)= (20) ^2=400
(46,186)
P (46) =208
(208-186)= (22) ^2=484
BI Calculation and statistic analytics Report

Introduction:
The aim of this report is to studying the relationship between Temperature and the number of passengers
that ride the main bus line in order to better serve their customer ,so the report include the statistical drill
down result based on these calculation so we will interpret the result based on them
The statistical analysis based on the temperature and passengers :
After we calculate the result of training, set MSE and RMSE and compare them with the validation test
we find that percentage of error in training set is less than validation set that are not include the training
set because they follow the linear Regression model and related to them based on the mean and the
standard deviation.
If the validation set better than training set we can tell that the linear regression model is perfect and
excellent for every data set insert to the model
If we look to each factor relate the model y (passengers) =slope +Intercept (Temperature)
Firstly the temperature in the table we find the highest temperature is 50 and the lowest temperature is 30
but there is redundant temperature in the table and each one of them the number of passengers is different
why? The difference between the passengers is very little that’s because the passenger may not satisfied
to enter the bus and the number of passenger high and the temperature is for example 46 so the
satisfaction of passenger is related to the number in the bus line
Secondly there is positive relationship between temperature and passenger and there are near to the range
1 so if they increase the temperature will increase and they near to the average and little far from the
average the standard deviation
finally the different between the predicted and the actual data is very high because there is wrong between
them and they predict based on the satisfaction and the daily temperature that they can predict it and find
using dashboard that show how many passenger we can carry in the bus
and if the passenger are not satisfied we can represent them as a performance visualization using KPIs to
make decision if the person went to the passenger by taking on his mind two factors :temperature and the
number of passenger if high the satisfaction will be low and as we see the temperature in the summer time
so all these factor when we calculate it and find the best prediction and prove that the data we made is
good enough to prove it
Conclusion
At the end this analysis is drilled down based on the environment of the situation and how we can solve it
if there is another way to make the data accurate and less of errors
BI Calculation and statistic analytics Report

You might also like