0
votes
0answers
2 views

SVD vs PCA for OLS

I know how to calculate PCA and SVD mathematically and I know that both can be applied to Linear Least squares regression. The main advantage of SVD mathematically seem to be that it can be applied ...
1
vote
0answers
1 views

Why does glmnet in caret give different predictions for different alphas even though lambda is zero?

In R, when using caret to train an elastic net regularization model, I find that different values of alpha give different predictions when the lambda parameter equals zero. This should not be the ...
0
votes
1answer
8 views

Suitable function to choose the best split in a regression tree/oblivious tree

My main objective is to construct a regression (decision) tree. It is a part of a boosting algorithm using additive regression trees. The first question is what other functions (other than least ...
1
vote
0answers
4 views

Glmnet Caret Package with small number of observations

I’m attempting to train a data set with 70 predictors, but only 35 observations with glmnet in the caret package. I’m trying to determine the best resampling method. It seems like 35 observations is ...
1
vote
3answers
7k views

How do I create subscales based on sets of items in SPSS

I have a 37-question scale, where each item belongs to one of four scales. The items for each scale are interspersed. How can I compute a scale score for each of the four scales? I need to be able ...
1
vote
0answers
8 views

glmnet: How to make sense of multinomial parameterization?

Following problem: I want to predict a categorical response variable with one (or more) categorical variables using glmnet(). However, I cannot make sense of the output glmnet gives me. Ok, first ...
0
votes
2answers
26 views

overfit gaussian regression

I have recently read a lot of posts about how maximizing the marginal likelihood in Gaussian Regression can cause overfitting. What is the best way out then? If we let the hyperparameters take values ...
0
votes
0answers
5 views

c-index for parametric links

I am conducting a binary regression using different sorts of parametric links (logistic, Pregibon, Aranda-Ordaz, ... see) and I would like to compare their predictive perfomance in a particular data ...
1
vote
1answer
667 views

Monte Carlo simulation exercise

The assignment is to run a simulation on excel to figure out the probabilities on obtaining a certain goal: John and Jane Doe are planning to save money to pay a house for their 6-month-old son, ...
3
votes
1answer
34 views

How can MANOVA report a non-significant p-value while LDA results in perfect separation of two groups?

I am new to statistics and currently got a dataset which contains $80$ dependent variables and $1$ independent variable with $2$ groups. MANOVA reports a $p$-value of $> 0.6$ on this dataset. But ...
0
votes
1answer
22 views

Parameter estimation in log linear models

Can anyone explain to me how parameter estimation is computed in log linear models? I followed this paper which is quite good, however I'm a bit confused in the parameter estimation part which is ...
13
votes
1answer
1k views

Detecting outliers in count data

I have what I naively thought to be a fairly straight forward problem that involves outlier detection for many different sets of count data. Specifically, I want to determine if one or more values in ...
4
votes
2answers
189 views

Best practice for dimensionality reduction with PCA and LDA: does it make sense to combine them?

Assume I have a dataset for a supervised statistical classification task, e.g., via a Bayes' classifier. This dataset consists of 20 features and I want to boil it down to 2 features via ...
6
votes
1answer
369 views

How is MANOVA related to LDA?

In several places did I see a claim that MANOVA is like ANOVA plus linear discriminant analysis (LDA), but it was always made in a hand-waving sort of way. I would like to know what exactly it is ...
1
vote
0answers
13 views

How does MATLAB compute the correlation coefficient for robust nonlinear least squares?

As part of a much larger project, I was running some fits and compared the results with what a colleague had and I noticed that our correlation coefficients weren't matching. So I have some data, and ...
1
vote
2answers
14 views

Time Series Anomaly Detection with Python

I need to implement anomaly detection on several time-series datasets. I've never done this before and was hoping for some advice. I'm very comfortable with python, so I would prefer the solution be ...
1
vote
1answer
72 views

Test whether triplet of numbers is significantly different from $K$ other triplets

Given $K$ triplets $t_k:=(a_k^{(1)},a_k^{(2)},a_k^{(3)}) \in \mathbb{R}^3, k=1,...,K$ and the triplet to test against them $t=(a^{(1)},a^{(2)},a^{(3)}) \in \mathbb{R}^3$. How can one test whether ...
0
votes
0answers
37 views

Estimating correlation hyperparameters of a Gaussian Process

I have an actual function that I need to simulate using a GP model. I've not done this before so I'm unclear of the steps. I have used the true function at different values of the inputs ($\vec X1, ...
0
votes
0answers
14 views

imputing missing values [on hold]

My work involves mining multivariate time series data and i am dealing with clinical dataset with unequal intervals. I am confused with what type of method to employ for handling missing values.
5
votes
2answers
108 views

What is the probability distribution of $1-\text{mean}(|A-B|)$ where $A$ and $B$ are independent U(0,1)?

I'm not well versed in statistics so I'm not sure if my question is worded exactly correctly but basically here's the problem I'm trying to solve: imagine you have two equal sized arrays of size n. ...
0
votes
1answer
210 views

How to define the multiplier range for variance test based outliers detection algorithm?

I have a variance test based outliers detection algorithm. The algorithm is exposed with a visual application where the user can configure its parameters namely the multiplier. The question is what ...
0
votes
2answers
25 views

Density plot narrow peaks and mean far away from peaks

I am trying to understand a density plot. It has narrow peaks and means are far away from the peaks. What would be the reason for such behavior? Should I be doing a different plot to understand this ...
0
votes
1answer
9 views

Convolutional neural network with non-image input data

Can CNNs be used with input data which is not an image? The reason I'm asking is because the original image is often clipped in size because of border effects when doing the convolution. But if the ...
1
vote
1answer
142 views

Statistical comparison of two signals

I need to develop an algorithm that will compare two signals and generate some metric(s) to describe changes between them. Signal processing and analysis isn’t my strong point so I would appreciate ...
0
votes
1answer
15 views

Running regressions on subsamples: Include the variable on which the subsample is based in the regressions?

Income is one of my independent variables in a sample of 2000 households. I'd like split the sample along household income lines (poor, vulnerable and non-poor). Is there a reason not to include ...
0
votes
1answer
55 views

Compare linear regression models (same and different response variable)

How can I (1) compare two linear models between years and (2) Can I compare 2 models with different response variables? My data have 4 variables: y_meas, x, year, y_calc. "y_meas" is a lab measured ...
1
vote
1answer
31 views

Removing collinear variables for LDA/QDA in R

I'm new to R and I've been searching for a while for a function which can reduce the number of explanatory variables in my lda function (linear discriminant analysis). Basically, I've loaded the ...
0
votes
2answers
52 views
+50

random forest variables importance with continuos and categorical variables and unbalanced output

I am a bit lost in the literature regarding the random forest importance. I am aware that there are different methods. I have a binary output variables where elements labelled with 0 are much ...
6
votes
5answers
6k views

How to identify structural change using a Chow test on Eviews?

I have this little problem and I would appreciate some help. As part of my master thesis, I have to identify a trend in a univariate (GDP) time series for different countries. I have to separate the ...
0
votes
0answers
10 views

How can I format the axis in my plot? [on hold]

I want to do a simple temperature over time curve in ggplot2. I have an average temperature every day over 8 months. When I plot this the x axis gets very crowded. How can I format the x axis so that ...
0
votes
0answers
4 views

Clustering multi-dimensional data with RapidMiner without knowing k

I am currently struggling to find an appropriate clustering for a data set with binominal label and 10 features (all of them continuous, but in different ranges). As far as handling the high ...
3
votes
1answer
129 views

If summarizing stats from multiple models is it meaningful to report a mean AIC?

I am currently summarizing results from several groups of models. Is it meaningful to report a mean AIC for each group of models? If not then how best to give a summary measure for each model ...
0
votes
0answers
15 views

What do I use to describe the difference between two groups in terms of found latent variables

I need to describe what the difference between two groups (patients and normal controls) consists of in terms of latent variables that I can describe within each group. For instance, given this PCA ...
1
vote
0answers
17 views

Poisson Process

I would appreciate a hint on this problem: A pedestrian wishes to cross a single lane of fast-moving traffic. Suppose the number of vehicles that have passed by time $t$ is a Poisson process of rate ...
0
votes
1answer
8 views

libsvm: scaling data results in less features?

I've scaled my training and testing data in BASH like so: ...
2
votes
1answer
183 views

Get fitted values estimated in ARIMA in Matlab

I am using Matlab to forecast time series data using ARIMA algorithm. I am able to get forecasted values, but unable to get the fitted values. This is what I mean. In R: ...
1
vote
3answers
175 views

Daily forecasting

We have three years of data for online visits at a daily level. We want to forecast the daily visits for the next 90 days. What would be the best method to capture weekday seasonality , holiday ...
4
votes
1answer
115 views

Why does this logistic GAM fit so poorly?

I am trying to create a logistic regression model with mgcv::gam with what I think is a simple decision boundary, but the model I build performs very poorly. A local regression model built using ...
2
votes
0answers
34 views

How to write up and report a Bayesian analysis?

Bayesian Estimation Supersedes the t-Test for John K. Kruschke is one of the most important papers that I had read explaining how to run the Bayesian analysis and how to make the plots. But the ...
0
votes
1answer
15 views

Sum of dependent R.V

I have two random variables whose PDF are parameterized by an unknown constant as follows: P(A;d) P(B;d) apparently, these two are not independent, so to find P(A+B;d) one cannot use convolution. ...
1
vote
1answer
254 views

Trying to use Holt-Winters to fit this data

I'm trying to fit the data in this message (daily temperatures) using the Holt–Winters technique in R, but can't get the seasonal example in here to work. Is this not possible with these data, or am I ...
0
votes
0answers
14 views

Standard error for multiple regression? [duplicate]

My understanding of the standard error of estimate for a simple linear regression as a function of the sum of squares over the number of observations, but how do we interpret the standard error for a ...
2
votes
3answers
433 views

Unable to provide random parameter with mlogit

I'm having trouble with rpar argument in the mlogit function (package mlogit). My dataset looks like this: ...
1
vote
1answer
44 views

Is there a name or reference in a published journal/book for the following variance formula?

Variance can be combined as $$v=\frac{1}{n-1}\left(\sum_{i = 1}^{numGroups}n_{i}(m_{i}-m)^2+ \sum_{i = 1}^{numGroups}(n_{i}-1)v_{i}\right)$$ where $v$ is the combined variance, $n$ is the total ...
0
votes
0answers
8 views

Intuitive explanation of the proximal operator

Can someone please give an intuitive explanation of the proximal operator? I understand that it is somehow related or is a generalization of projection, but other than that, I'm unable to grasp it.
0
votes
0answers
11 views

How do I choose the right classifier?

Broadly speaking, is there a good tutorial, possibly online, about which classifier is good for what, depending on number of samples and other statistical properties of the input data?
0
votes
0answers
10 views

How to normalize a severly positively skewed data for multiple regression [on hold]

I have a severely positively skewed data set that I need to normalize without significantly reducing my valid case number. I have tried log, log+1, square and inverse, without success. I have 65% of ...
0
votes
0answers
31 views

Test for autocorrelation

I wanted to test if there's significant autocorrelation in my data. Here's the reproducible code(R!) and the result. It looks like that dwtest and bgtest and acf are all too much different. Can ...
1
vote
0answers
29 views

What is relationship between Fisher Information and Variance in natural exponential Family?

I know that $Var(\hat\theta)\geq 1/I(\theta)$ where $I(\theta)$ is Fisher information. Let take an example of natural exponential family with density $f(x)=\lambda\exp(-\lambda x)$. In this case we ...
1
vote
1answer
62 views

How to calculate the confidence interval for a combination of a proportion and continuous data

I want to calculate the 95% confidence interval for a combination of binary data and continuous data. Lets say I want to estimate the production of a certain item from all factories of a given ...

15 30 50 per page