Re: normality and regression analysis

Mike wrote in message 8ffek1$1q2$[EMAIL PROTECTED]... I would like to obtain a prediction equation using linear regression for some data that I have collected. I have read in some stats books that linear regression has 4 assumptions, 2 of them being that 1) data is normally distributed and 2)

Benford's law and simulation

Background: Theodore Hill showed, in a paper published in Statistical Science 1995, that if sequences of random variables $\{X\sb n\}$ are selected at random in a scale (base) unbiased way, then the mantissa distributions of the combined sample will converge to Benford's law---a random

Re: normality and regression analysis

Hi Mike. For the most popular linear regression Ordinary least squares (OLS), you also need to have your X variable (i.e. the independent variable) having a relatively small error. Your initial work suggests large-ish error in both variables with non-normal error structure. This makes things a

Re: Regression Through the Origin

1. Are you sure the coefficients for x and x^2 are the same ? 2. If there is no intercept, it is unclear that R^2 is meaningful. 3. The RSSQ (and thus the MSE) must be smaller if the constant is included. At 12:44 + 05/12/2000, Homie wrote: I am currently running a simple quadratic

RE: normality and regression analysis

Mike...regression assumptions are more concerned with distributional characteristics of the errors than the actual raw score, in that if residuals are normally distributed, there is a constancy of variation of the errors across the x axis (i.e., homescedasticity), etc., then non-normality

Re: Regression Through the Origin

On Fri, 12 May 2000 12:44:05 GMT, **[EMAIL PROTECTED] (Homie) wrote: I am currently running a simple quadratic model, with and without the constant (y=a+bx+bx^2) and (y=bx+bx^2). The model fit is much better when the constant is not included (see below). Although there is some minor

Re: normality and regression analysis

In article 8ffek1$1q2$[EMAIL PROTECTED], Mike [EMAIL PROTECTED] wrote: I would like to obtain a prediction equation using linear regression for some data that I have collected. I have read in some stats books that linear regression has 4 assumptions, 2 of them being that 1) data is normally

Re: normality and regression analysis

In reply to Mike's question Allan makes the important point: There is absolutely no requirement that the predictors (or independent variables) should have a normal distribution, in fact the opposite. Ideally, the predictors should be from a designed experiment and hence will not even be

Sinopsis

Dear list members, is there any guide, book, rules or norms that indicates/suggests "what" and "how" to put statistical information into a condensed publication like a synopsis or a census publication? We we are dealing here with educational data. Thanks

Re: Correlation over time

In my former life as a neurobiologist, we analysed the relationship between rodent investigative sniffing and the limbic theta rhythm. In short, rodents tend to exhibit a preferred phase relationship between these two signals, both of which run about 5-9 Hz. We analysed short (1-2 sec) epochs.