In: Statistics and Probability
Simple linear regression, like ARIMA, involves statistical modeling. Unlike decomposition, averaging and smoothing methods, fitting a simple linear regression model to data involves statistical inference.
Moreover, several assumptions/conditions need to be satisfied in order to use a simple linear regression model. One might think that this added level of complexity would make regression analysis less likely to be used in practice. On the contrary, it is widely used by management. Why do you suppose this is the case? What advantages does simple linear regression have over the forecasting methods we've covered so far? Can you give an example of how simple linear regression may be used in your area of employment and/or expertise?
Why do you suppose this is the case ?
Ans -
Simple Linear Regression is the method which needs to hold several
assumption in order to use it.
These assumptions must satisfied in order to use the method
effectively and to reduce the data anomalies
such as outlier, constant variance throughout the observations and
normality of the errors.
If we do not consider these assumptions then also we can use these
methods but the model will not remain robust.
Once new data we put it the model it will deviate from its original
behavior.
Hence it is so popular in its use even it has many assumptions to
hold.
What advantages does simple linear regression have over the other forecasting method?
Ans-
The Simple Linear Regression method is one of the simplest method
to forecast/predict the data points.
It is also easy to interpret and meaning full inferences can be
drawn from this model.
Leymann can also understand this basic model. Also, it is available
in all the softwares and statistical packages.
These are some of the advantages of this model.
Example of simple linear regression in employment area -
Ans -
Employees salary is dependent on the skill sets and prior
experiance.
This is how we are using this model.