I have trained an linear SVM which takes a pair of objects, computes features and is expected to learn a semantic similarity function between objects(we can say that it predicts whether the two objects are similar enough that they should be merged or not). The problem I am facing is that the predictions can be … Read more
I have some data which follow a nonlinear relationship like that displayed in the plot below. The non-linear data do not come directly from the explicit function written in the code below! x1 <- seq(-1,-0.0001,len=500) x2 <- seq(0.0001,1,len=500) df <- data.frame(x=c(x1,x2),y=c(1-0.0001^x1,0.0001^(-x2)-1)) plot(df[,1],df[,2],type="p") Given that there are negative values in both x and y, how do … Read more
I have some data which follow a nonlinear relationship like that displayed in the plot below. The non-linear data do not come directly from the explicit function written in the code below! x1 <- seq(-1,-0.0001,len=500) x2 <- seq(0.0001,1,len=500) df <- data.frame(x=c(x1,x2),y=c(1-0.0001^x1,0.0001^(-x2)-1)) plot(df[,1],df[,2],type="p") Given that there are negative values in both x and y, how do … Read more
In this question a commenter says that "differencing a series that is not integrated is certainty problematic from the statistical perspective". What is an integrated time series, and why is differencing a series that is not integrated problematic? Best Answer Consider the first difference $Delta u_t$ of a linear process (a fairly general way of … Read more
I am running an experiment which measured using Likert Scales and I have 6 variables out of 24 which either have significant skewness or kurtosis. These are mixed with some positive and some negative values. I transformed my data using log transformations, Square root transformations, reciprocal transformations and reverse score transformations, but this did not … Read more
Machine learning (ML) uses linear and logistic regression techniques heavily. It also relies on feature engineering techniques (feature transform, kernel , etc). Why is nothing about variable transformation (e.g.power transformation) mentioned in ML? (For example, I never hear about taking root or log to features, they usually just use polynomials or RBFs.) Likewise, why don't … Read more
Machine learning (ML) uses linear and logistic regression techniques heavily. It also relies on feature engineering techniques (feature transform, kernel , etc). Why is nothing about variable transformation (e.g.power transformation) mentioned in ML? (For example, I never hear about taking root or log to features, they usually just use polynomials or RBFs.) Likewise, why don't … Read more
In meta-analysis, it seems a standard practice to take the natural log of the response ratio before evaluating it. My question is why? That is, if I have a treatment mean (Xe) and a control mean of (Xc), and the response ratio (RR) is defined as Xe/Xc, why would I take the natural logarithm of … Read more
In meta-analysis, it seems a standard practice to take the natural log of the response ratio before evaluating it. My question is why? That is, if I have a treatment mean (Xe) and a control mean of (Xc), and the response ratio (RR) is defined as Xe/Xc, why would I take the natural logarithm of … Read more
I have data for which I would like to take the log transformation before doing OLS. The data include zeros. Thus, I want to do a log(x + c). I know a traditional c to choose is 1. I am wondering whether there is a way to have the data choose c such that there … Read more