Method of least squares in reverse order: fitting of linear curve to average minimum temperature data at Guwahati and Tezpur

In 2015, Rahman and Chakrabarty have developed one method by stepwise application of the principle of least squares of estimating parameters associated to a linear curve, based on the principle of elimination of parameter(s) first and then minimization of sum of squares of errors in place of the principle of least squares namely minimization of sum of squares of errors first and then elimination of parameter(s), where all values of the ratio of the difference of each pair of observed values of the dependent variable to the difference of each pair of observed values of the independent variable have been used in estimating parameters involved in the curve. In this paper, discussion has been made on how parameters can be estimated using the independent values from among the values of the said ratio. One numerical application of the method has also been discussed in fitting of linear regression of monthly mean minimum temperature on the monthly average length of day at Guwahati and also at Tezpur.