The values of the slope and intercept are calculated using maximum likelihood estimation using the least squares method." that minimizes the sum of the squared errors of the distances between all points and lines. The ``yi'' in maximum likelihood estimation means ``most likely''. In other words, roughly speaking, it means ``estimating the most plausible value.'' The "most plausible value" is the value that is expected to appear most frequently in terms of probability (a little difficult for me as an outsider...).
Let's consider a diagram to understand why it is necessary to square the error in the distance between a point and a line. In the case of this function, a person who
Chinese Student Phone Number List studied for 2 hours would get approximately 150 points. On the other hand, people with score R studied for 2 hours and got 330 points. In other words, there is an error of 330-150=180 from point R to the line (function). Also, those who studied for 3 hours scored around 190 points on the function.
However, people with score S only got around 110 points. The error at this time is 110-190=-80. It will be a negative value. If there are positive and negative values, they will cancel each other out when calculating the sum. Therefore, by squaring the error, you can avoid thinking about pluses and minuses. Find using Python Mr. Nakajima: "You can use ``LinearRegression'' in the Python machine learning library ``scikit-learn'' to find the slope and intercept.
Zaloguj
Zarejestruj
FAQ
Szukaj
Użytkownicy
