statsmodels

I have been using ECDF (empirical cumulative distribution function) from statsmodels.distributions to plot a CDF of some data. However, ECDF uses a step function and as a consequence I get jagged-looking plots. So my question is: Do scipy or statsmodels have a ECDF baked-in without a step function? By the way, I know I can do this: hist, bin_edges = histogram(b_oz, normed=True) plot(np.cumsum(hist)) but I don't get the right scales. Thanks! If you just want to change the plot, then you could let matplotlib interpolate between the observed values. >>> xx = np.random.randn(nobs) >>> ecdf = sm

Say I have two random variables: X ~ Beta(α1,β1) Y ~ Beta(α2,β2) I would like to compute distribution of Z = XY (the product of the random variables) With scipy , I can get the pdf of a single Beta with: from scipy.stats import beta rv = beta(a, b) x = np.linspace(start=0, stop=1, num=200) my_pdf = rv.pdf(x) But what about the product of two Betas? Can I do this analytically ? (Python/Julia/R solutions are fine). Sven Hohenstein For an analytical solution, have a look at this paper and this answer . A numerical approach in R set.seed(1) # for reproducability n <- 100000 # number of random

Closed. This question is off-topic . It is not currently accepting answers. Want to improve this question? Update the question so it's on-topic for Stack Overflow. Closed 3 years ago . I know there is an weighted OLS solver , and a constrained OLS solver . Is there a routine that combines the two? You can simulate OLS weighting by modifying the X and y inputs. In OLS, you solve β for X t X β = X t y . In Weighted OLS, you solve X t X W β = X t W y . where W is a diagonal matrix with nonnegative entries. It follows that W 0.5 exists, and you can formulate this as (X W 0.5 ) t (XW 0.5 ) β = (X W

I'm trying to run a logistic regression in statsmodels on a large design matrix (~200 columns). The features include a number of interactions, categorical features and semi-sparse (70%) integer features. Although my design matrix is not actually ill-conditioned, it seems to be somewhat close (according to numpy.linalg.matrix_rank , it is full-rank with tol=1e-3 but not with tol=1e-2 ). As a result, I'm struggling to get logistic regression to converge with any of the methods in statsmodels. Here's what I've tried so far: method='newton' : Did not converge after 1000 iterations; raised a