I am using the kernlab package for svm in R.I am using the linear kernel so that I can directly check the importance of the feature vectors, that is my variables.Using the coefficients of these feature vectors,I am required to calculate the weight of the various factors in the model,so that the linear separating plane that the svm will draw in my feature space can we evaluated. Basically I want to calculate the w in transpose(w)*x + b. Could someone please suggest what is to be done. I used the fields alpha and b and apha index and tried to logically calculate the weight vector, but to verify if I was calculating correctly I tried to predict on a test sample its correct predicted score, and this did not match the value predicted by the inbuilt predict function. How to calculate the weights?
Related
When using the GAMLSS package in R, there are many different ways to fit a distribution to a set of data. My data is a single vector of values, and I am fitting a distribution over these values.
My question is this: what is the main difference between using fitDist() and gamlss() since they give similar but different answers for parameter values, and different worm plots?
Also, using the function confint() works for gamlss() fitted objects but not for objects fitted with fitDist(). Is there any way to produce confidence intervals for parameters fitted with the fitDist() function? Is there an accuracy difference between the two procedures? Thanks!
m1 <- fitDist()
fits many distributions and chooses the best according to a
generalized Akaike information criterion, GAIC(k), wit penalty k for each
fitted parameter in the distribution, where k is specified by the user,
e.g. k=2 for AIC,
k = log(n) for BIC,
k=4 for a Chi-squared test (rounded from 3.84, the 5% critical value of a Chi-squared distribution with 1 degree of fereedom), which is my preference.
m1$fits
gives the full results from the best to worst distribution according to GAIC(k).
I know when random forest (RF) is used for classification, the AUC normally is used to assess the quality of classification after applying it to test data. However,I have no clue the parameter to assess the quality of regression with RF. Now I want to use RF for the regression analysis, e.g. using a metrics with several hundreds samples and features to predict the concentration (numerical) of chemicals.
The first step is to run randomForest to build the regression model, with y as continuous numerics. How can I know whether the model is good or not, based on the Mean of squared residuals and % Var explained? Sometime my % Var explained is negative.
Afterwards, if the model is fine and/or used straightforward for test data, and I get the predicted values. Now how can I assess the predicted values good or not? I read online some calculated the accuracy (formula: 1-abs(predicted-actual)/actual), which also makes sense to me. However, I have many zero values in my actual dataset, are there any other solutions to assess the accuracy of predicted values?
Looking forward to any suggestions and thanks in advance.
The randomForest R package comes with an importance function which can used to determine the accuracy of a model. From the documentation:
importance(x, type=NULL, class=NULL, scale=TRUE, ...), where x is the output from your initial call to randomForest.
There are two types of importance measurements. One uses a permutation of out of bag data to test the accuracy of the model. The other uses the GINI index. Again, from the documentation:
Here are the definitions of the variable importance measures. The first measure is computed from permuting OOB data: For each tree, the prediction error on the out-of-bag portion of the data is recorded (error rate for classification, MSE for regression). Then the same is done after permuting each predictor variable. The difference between the two are then averaged over all trees, and normalized by the standard deviation of the differences. If the standard deviation of the differences is equal to 0 for a variable, the division is not done (but the average is almost always equal to 0 in that case).
The second measure is the total decrease in node impurities from splitting on the variable, averaged over all trees. For classification, the node impurity is measured by the Gini index. For regression, it is measured by residual sum of squares.
For further information, one more simple importance check you may do, really more of a sanity check than anything else, is to use something called the best constant model. The best constant model has a constant output, which is the mean of all responses in the test data set. The best constant model can be assumed to be the crudest model possible. You may compare the average performance of your random forest model against the best constant model, for a given set of test data. If the latter does not outperform the former by at least a factor of say 3-5, then your RF model is not very good.
We wanted to see whether the recorded EEG data would be a good classifier to classify participants into group A or B.
I used the LDA function in the MASS package in R for a linear discriminant analysis. To make sure the model would be generalizable, we added a leave-one-out cross validation (CV = TRUE). All of this worked very well.
Now I would like to visualise this classification based on the discriminant scores. More specifically, show both groups on the x-axis and the discriminant scores on the y-axis. Something like this: classification based on discriminant scores
However, when applying CV = TRUE, it is not possible to use the predict function on the LDA object to generate LD1 values. Also, no coefficients of linear discriminants are returned (which could be used to calculate the discriminant scores myself (based on these links: What are “coefficients of linear discriminants” in LDA? and Discriminant functions)).
When I remove the CV = TRUE, I can use the predict function afterwards and get the LD1 values (i.e. discriminant scores) needed for the visualisation. However, my classification results differ depending on whether or not I apply the leave-one-out cross validation. For example, when using CV = TRUE, a lower number of participants would be classified in the correct group. So if I would plot the data based on the results of the linear discriminant analysis without leave-one-out cross validation, it would not correspond to the results I would describe.
My question is whether there is a way to get the LD1 values or the coefficients of linear discriminants when applying CV = TRUE to the LDA function of the MASS package?
Or if someone knows another option to produce a plot similar to the one I added above?
I'm a chemist and about an year ago I decided to know something more about chemometrics.
I'm working with this problem that I don't know how to solve:
I performed an experimental design (Doehlert type with 3 factors) recording several analyte concentrations as Y.
Then I performed a PCA on Y and I used scores on the first PC (87% of total variance) as new y for a linear regression model with my experimental coded settings as X.
Now I need to perform a leave-one-out cross validation removing each object before perform the PCA on the new "training set", then create the regression model on the scores as I did before, predict the score value for the observation in the "test set" and calculate the error in prediction comparing the predicted score and the score obtained by the projection of the object in the test set in the space of the previous PCA. So repeated n times (with n the number of point of my experimental design).
I'd like to know how can I do it with R.
Do the calculations e.g. by prcomp and then lm. For that you need to apply the PCA model returned by prcomp to new data. This needs two (or three) steps:
Center the new data with the same center that was calculated by prcomp
Scale the new data with the same scaling vector that was calculated by prcomp
Apply the rotation calculated by prcomp
The first two steps are done by scale, using the $center and $scale elements of the prcomp object. You then matrix multiply your data by $rotation [, components.to.use]
You can easily check whether your reconstruction of the PCA scores calculation by calculating scores for the data you input to prcomp and comparing the results with the $x element of the PCA model returned by prcomp.
Edit in the light of the comment:
If the purpose of the CV is calculating some kind of error, then you can choose between calculating error of the predicted scores y (which is how I understand you) and calculating error of the Y: the PCA lets you also go backwards and predict the original variates from scores. This is easy because the loadings ($rotation) are orthogonal, so the inverse is just the transpose.
Thus, the prediction in original Y space is scores %*% t (pca$rotation), which is faster calculated by tcrossprod (scores, pca$rotation).
There is also R library pls (Partial Least Squares), which has tools for PCR (Principal Component Regression)
I would really appreciate any help with specifying probability weights in R without using the Lumley survey package. I am conducting mediation analysis in R using the Imai et al mediation package, which does not currently support svyglm.
The code I am currently running is:
olsmediator_basic<-lm(poledu ~ gateway_strict_alt + gender_n + spline1 + spline2 + spline3,
data = unifiedanalysis, weights = designweight).
However, I'm unsure if this is weighting the data correctly. The reason is that this code yields standard errors that differ from those I am getting in Stata. The Stata code I am running is:
reg poledu gateway_strict_alt gender_n spline1 spline2 spline3 [pweight=designweight]).
I was wondering if the weights option in R may not be for inverse probability weights, but I was unable to determine this from the documentation, this forum or elsewhere. If I am missing something, I really apologize - I am new to R as well as to this forum.
Thank you in advance for your help.
The R documentation specifies that the weights parameter of the lm function is inversely proportional to the variance of the observations. This is the definition of analytic weights, or aweights in Stata.
Have a look at the ipw package for inverse probability weighting.
To correct a previous answer - I looked up the manual on weights and found the following description for weights in lm
Non-NULL weights can be used to indicate that different observations have different variances (with the values in weights being inversely proportional to the variances); or equivalently, when the elements of weights are positive integers w_i, that each response y_i is the mean of w_i unit-weight observations (including the case that there are w_i observations equal to y_i and the data have been summarized).
These are actually frequency weights (fweights in stata). They multiply out the observation n number of times as defined by the weight vector. Probability weights, on the other hand, refer to the probability that observations group is included in the population. Doing so adjusts the impact of the observation on the coefficients, but not on the standard errors, as they don't change the number of observations represented in the sample.