All Questions

0
votes
0answers
2 views

How to prove that X'ε=0 in regression models?

If X is a nxn matrix of some data exogenous to and ε is a 1xn matrix of residuals that sum up to zero, why is X'ε=0?
0
votes
0answers
2 views

Biased viterbi training result

I try to use GMM-HMM model to infer the topic of sentences in a short paragraph. While instead of using normal Baum-Welch optimization, I use viterbi training as follows. I use average of word ...
0
votes
0answers
4 views

Can't sample all conditions for each participant — strategies to compensate?

I am attempting to design a study using Category Comparison Rating / Double Stimulus Comparison Scale to evaluate participant perception of different stimulus conditions. The original study design ...
0
votes
0answers
2 views

metafor meta-analysis for pre-post differences without a control group

I am conducting a meta-analysis of the effectiveness of a pain medication. The studies I included do not have a control group. I am used the following commands but by reading others comments I ...
0
votes
0answers
8 views

How to do a simulation for hypothesis testing in R? [on hold]

This is my data set called native. This data set has 500 rows of data, but here I've displayed only 9 rows of data. ...
1
vote
0answers
7 views

What is the relation between a loss function and an energy function?

A loss function is a function that measures the distance between the expected value and the actual value of a model (an example of a loss function is the cross entropy). An energy function can be ...
0
votes
1answer
7 views

Calculating a continuous variable with regression trees

I have sample records with several attributes (predictors) and a predicted variable Yes/No. What I need is, given new data that omits the column Yes/No, to know what is the probability of Yes. Note ...
0
votes
0answers
12 views

Paired t-test means that the variances of the 2 samples are the same?

Just a question, not about coding. Wanted to know if paired samples imply that they have the same variance. Thank you.
0
votes
0answers
10 views

Distribution of the product of $n$ i.i.d log-normal random variables [duplicate]

Let $X_1,\ldots,X_n$ be independent log-normal random variables such that $$log(X_i)\sim N(\mu,\sigma^2)\ \ \forall i=1,\ldots,n$$ Then what can be said about the distribution of the random variable $...
0
votes
0answers
6 views

Minimum sample size for time series cross-validation (tsCV)

I am doing cross-validation of an autoregressive neural network model and I am using the tsCV function (forecast package) ...
0
votes
0answers
7 views

CNN, “squared” or "non-squared image?

I'm working on a project about image recognition. In my dataset I have images of different size, all rectangular image (the most 640x480 and 1280x640). I would like to build my classifier to ...
0
votes
0answers
4 views

Multi-level factor design of experiments

How should I select interactions in order to be able to encode multi-level features using interactions of main features in my design? I'm wanting to create an experimental design to be used to create ...
0
votes
0answers
9 views

Design of Experiments - Minimum Number of Experiments

I am new to experimental design. I am currently trying to determine the minimum number of trials required to estimate a linear regression model with one response variable and three predictive ...
1
vote
0answers
16 views

Does bias in regression coefficients affect the prediction?

Suppose i have biased coefficients in an ols estimation because of omitted variable bias or sample selection bias or something else that produce biased coefficients. If the goal is to get a model for ...
0
votes
0answers
9 views

Nominal variables analysis

I would like to answer the following question by using statistical analysis: What are differences and similarities in CSR practices between headquarters(hq) and subsidiaries(sub) in developing and ...
0
votes
0answers
6 views

Consistency of Jackknife and Bootstrap Estimates for Simple Statistics

For both, Jackknife and the simple Bootstrap I'm supposed to proof the consistency of the variance and bias estimators. (Only for the simple case, that our statistic is a function of the sample mean.) ...
0
votes
0answers
12 views

Rounding numbers in a $\chi^2$ test

I wanted to ask about rounding numbers in my $\chi^2$ test for discrete real-world values. Sadly, teacher wants to see if I was tossing my coin fairly (and by my own hand) for a previous statistics ...
1
vote
0answers
26 views

OLS as estimator

we've been given following question, but have some trouble getting started, can anyone help out? $\pi_{t}=\alpha_{1}+\alpha_{2} u_{t}+\alpha_{3} \pi_{t+1 | t}^{e}+\eta_{t}, \quad t=1,2, \ldots, T$ ...
2
votes
1answer
15 views

Are the standard errors in the coefficients of a linear regression corellated?

Lowly physical scientist here so please excuse my ignorance. I have a data set of thermal expansion coefficient, $\alpha$, against temperature, $T$ that I fit with a linear regression. From this, I ...
2
votes
0answers
11 views

80 pick 20, sort asd, sum per 4 numbers, why result a bit bias towards even number

I am working on the probabilities of a game where 1:they pick 20 unique number(non-repeating) from range 1 to 80 2:sort those 20 number in ascending order 3:sum every 4 number, take the last digit ...
0
votes
0answers
8 views

When is the bias of a statistic of the form a/n + b/n^2 + c/n^3 +

In many books the bias-correction of the Jackknife resampling method is being prooved under the assumption, that the bias has a special form, namely a/n + b/n^2 + c/n^3 * ... Sometimes it's written "...
0
votes
0answers
8 views

why can you perform t-test with the critical values of normal distribution? [duplicate]

I have estimated parameters and I want to check if they are significantly different from zero by using t-test. But since I have a large sample, I know I can compare the values with the critical values ...
-1
votes
0answers
4 views

NGramTokenizer not working in R Code [on hold]

install.packages("readr") install.packages("RWeka") install.packages("tm") install.packages("SnowballC") R code is: Tokenizer<-function(x) NGramTokenizer(x,weka_control(min=2,max=2)) head(...
0
votes
0answers
3 views

Adjusting weights in meta-analyses using R

I am conducting a meta-analysis in R. Some of my studies provide more than one effect size for the same sample. I would like to account for this by adjusting the weight that each within-study effect ...
1
vote
1answer
18 views

How does fisher.test calculate the confidence interval for the odds ratio in R?

The fisher.test function in base R by default returns a confidence interval for the odds ratio in a 2x2 contingency table. For example: ...
0
votes
0answers
6 views

classify samples from two different gaussian distributions [duplicate]

Assume we have two sampling process, i.e. we can draw samples from two different gaussian distributions P and Q. At first, we draw samples from gaussian distribution P only. But after some time t, we ...
0
votes
0answers
14 views

Why do the PCs of PCA seem to be correlated? [on hold]

If even the first PCs seem to be highly correlated what could be the problem? Does it have to do with the data - maybe their "structure"?
0
votes
0answers
8 views

Highly significant t-test statistics, any way to correct this?

I have been performing several paired t-test, where I test whether the sample means are different between two groups. For all of my tests (approx. 30), which all carries 13-14,000 observations, I ...
0
votes
1answer
10 views

Training Perceptrons with Backprop

Is it possible to train a simple perceptron with a threshold activation function such as this one: https://en.wikipedia.org/wiki/Perceptron with Backpropagation instead of the perceptron rule? is it ...
0
votes
1answer
11 views

Scoring rules for count models on: training data vs. validation data

In order to evaluate and compare count models (e.g. Poisson regression), we can calculate scoring rules (e.g. Brier Score, Dawid-Sebastiani score, etc.) which are explained here: Error metrics for ...
0
votes
0answers
11 views

Softmax where the max probability is less than one

I have a neural network with a softmax at the end. Something like this: ...
2
votes
1answer
12 views

Can we calculate a confidence interval using a permutation testing approach?

Tim Hesterberg's "What Teachers Should Know about the Bootstrap" is excellent. On page 77 he writes: "We cannot use permutation testing to obtain confidence intervals." Is that correct? If so, can ...
0
votes
0answers
10 views

Gaussian mixture model (GMM) to cluster high dimensional dataset

My data set has ~20.000 dimensions. I want to use "Gaussian mixture model " to cluster them. To Construct mixtures, multivariate distributions are required. the multinomial distribution is the ...
0
votes
0answers
14 views

Defining a covariance matrix [on hold]

I'm working with a Gaussian Process, and my covariance matrix is isotropic, i.e. it is defined just by the distance of the points locations. Suppose I'm working with Squared Exponential Covariance. ...
0
votes
0answers
12 views

why the validation accuracy of deep CNN is high but not stable

I'm training a CNN on some vibration data, and I get some somewhat strange results. I found the validation accuracy is unstable. In some epoch, the val_acc may be 90%+, but in the next epoch, it may ...
0
votes
0answers
4 views

using tf.metrics [on hold]

I'm trying to get AUC metrics for a keras model, the API docs don't explain this method in much detail ...
0
votes
2answers
22 views

Covariance of two normally distributed variables

I saw in a statistic book that "It can be prooved that if two normally distributed variables have covariance = 0, they are independent". How can I start this proof? Can I say that $cov(X,Y) = E(XY) ...
0
votes
1answer
13 views

Using different color scales for correct and incorrect predictions in an sklearn confusion matrix

I'm trying to illustrate the accuracy of my neural net predictions using a matplotlib plot of an sklearn confusion matrix. My problem is that the size of my test set is quite large the accuracy is ...
-1
votes
1answer
10 views

Select relevant data to feed model

I am building a regression algorithm to predict the cost of a product defined by several features. I use a supervised rule-based model to do the predictions. The dataset is based on former ...
-1
votes
0answers
18 views

Pearson correlation calculation [on hold]

I need to calculate the Pearson correlation coefficient for 2 different variables, easy, but with time series. I have 2 different variables (68 in each) data from 2009 to 2018. What would be the ...
1
vote
3answers
110 views

Calculate variance the right way with two random variables

I'm currently assigning a introductory stats class, and I just can't seem to find out when to use the different variance identities. I have provided an example of an assignment where I got it wrong, ...
0
votes
0answers
6 views

Standard error clustering in Stata redhdfe and R felm

I am trying to replicate Statas reghdfe in R. in Stata: ...
0
votes
1answer
13 views

Compare values within the single variable

I have a 5 categories of the occupations — starting from jobs with low physical effort like an IT guy, to the labor-intensive occupations like coal-mining etc. The categories are numbered from 1 to 5. ...
0
votes
0answers
6 views

Interpreting Dprime values

I had an unbalanced design in my experiment, therefore instead of running logistic regression models for the analysis I calculated Dprime values for each experimental condition and then took these d' ...
0
votes
0answers
5 views

How to avoid overfitting by entities in short text classifcation?

I am binary classifying headlines. This headlines are between 1 to 7 words, and sometimes include the name of the person who created them or locations. I am able to classify with 81% precision but ...
0
votes
0answers
6 views

Find number of elements to check for finding false negative rate

I have 3500 pictures that were classified in four different possibilities of match: very low (16%), low (1%), high (30%), very high (53%) All labels are not known, i.e., I don't know if they were ...
0
votes
0answers
10 views

Social Network Like Prediction

I'm currently using RapidMiner to predict the number of likes of the posts according to certain training variables (ID, number of likes, hour, month, weekday, type of post (Photo, Video, Status & ...
3
votes
0answers
26 views

Can two random variables be independent in some basis and dependant in other?

If some random variables forming N dimensions are dependant on each other is it possible that in a different coordinate system they'd be independent? For example if (X, Y) are two dependent RVs is it ...
0
votes
0answers
8 views

simr power simulation testing interaction term for 2 factors

I have a model with two factors: a (4 levels) and b (3 levels). Each participant receives two problems (id as a random effect). I want to estimate sample size based on pilot data to give me ...
0
votes
0answers
6 views

Can normalized values and original values be combined in a feature vector classification?

I standardized feature vector before SVM classification in MATLAB. The feature vector consists of time domain signal features and its normalizations i.e., feature vector is a combination of actual ...

15 30 50 per page