Search Results

It is not assumed that the "explicative variables" must be independent from each other. However, we must be aware of the problem of colinearity. If your explicative variables are correlated, then the …

I don't know aything about how the relaimpo package works and neither am I sure about what you exactly mean by "relative importance of regressors", but you must definitely include every "non-significa …

What you have in your hands looks like a classification problem rather than a regression one. … Fortunatelly, you already know a classification technique (confusingly enough: logistic regression) You may also want to tree different classifiers such as decision trees, which work great with categorical …

A good measure for how well 2D data fit into a line would be the determination coefficient. As you are familiar with R, you may want to compare compare: a=rnorm(100) b=rnorm(100) plot(a,b) cor(a,b)^2 …

I know no Python so I don't really know what you're talking about. Is there a possibility to build the feature stepwise? This means, first you build the model with all the features, then compare tha …

First, for there being a regression, there should be parameters! … I will assume 3000 and 1 are in this case It is linear regression if you consider your "employee age squared" as a variable, so, strictly speaking, it is a linear regression only after a transformation …

Let $y=x^2$ A linear model will not be able to capture anything and will just return $\beta_0$ as the mean and $\beta_1=0$ However, a regression tree will find a split based on the value of $x$ My point …

It really depends on the amount of data you have and how "precise" it is (i.e. what the real difference between, let's say, 70 and 71 actually is, or is the difference between 60 and 70 really the sam …

Linear regression assumes all your observations are independent from each other. … Also, in linear regression we study how other predictors ("regressors") influence our target value. …

Your expression is a formula for the standard estimation error for a given parameter $\beta$ in terms of the sample mean, the observations. $\sigma$ stands for the standard deviation of the errors $ …

In case of simple regression, it's just the square of the correlation coefficient between $Y$ and $X$) The adjusted $R^2$ is the same thing but compensating for the number of parameters (theoretically, … Adjusted $R^2$ is useful when comparing models with different number of parameters, so in simple regression we don't really care too much) The final line is a test on whether every parameter $\beta$, …

If Rank and Income are independent, you can do as follows: Top 15% observations have a 1 in 3 chance of being in the top 20% but not in the top 10% and a 2 in 3 chance of actually being in the top 10 …

I think you are right: regression (linear or otherwise) may be the way to go. …

Your model can still be fitted with colinear features, there is no "a priori" problem with that. The issue begins when you try to simplify to determine which predictors are having the most influence …

"It is known that increasing the complexity of a regression or classification model reduces the training error". So... it is trivial! Assume model A is more complex than model B. …