0
votes
0answers
4 views

Improve a tagging system based on future data learning

I need to improve a system that tags events as good/bad for model training set creation. The current system relies on 1-2 data points following the event (for the same customer) that could happen ...
0
votes
0answers
7 views

ggplot2: timeseries scale_x_date breaks [on hold]

When I use breaks with scale_x_date, it no longer respects the limits of the data, or the limits= parameter. Consider: ...
0
votes
1answer
17 views

Interpreting stripe like shapes on funnel plot from meta-analysis

I have a set of 130 measurements from about 30 different studies. I have made a random effects meta-analysis of SMD and ROM. I used the DL method. I also made funnel plots to check for publication ...
3
votes
1answer
86 views

Transform normal distribution to skewed distribution without changing its support

I've found many questions and answers about transforming skewed distribution to normal. This question might arise because the simplicity of working with normal data. But, is there any function that ...
0
votes
0answers
7 views

Correlation is moderate but Path coefficients (in pls-sem) show insignificant results

I am performing PLS-SEM analysis (usng plspm package), the outer loading's are significant as CI do not include zero on any of them However, large number of path coefficient's (i.e, inner model)are ...
0
votes
0answers
9 views

Random Variable Decomposition Standard Error

I have a decomposed random variable $X$ into partitions $A_1,A_2,\dots A_m$. I know how to compute the expected value of X and the variance of X given the variance and the standard errors of $X$ ...
0
votes
1answer
20 views

Simulate ARIMA Model in R using same starting values as original time series?

I have built an ARIMA model in R with the forecast package's auto.arima() function. I want to simulate the ARIMA model with the same starting values as the original time series. For example, if my ...
0
votes
0answers
7 views

Structural Equation Modeling [on hold]

How can we develop Structural Equation modeling ?E.g. Scientific Model.in so many papers i have observed it.basically Emerald publications and one more thing,which are the software that can help me ...
0
votes
0answers
5 views

Regarding pattern recognition and forecasting next set of sequence [on hold]

I have been working on use case where I have some sample data as below : Date Device Name Time Facility Code 1/6/2015 Tablet X 16:18:48:168 22 A 1/6/2015 Tablet X 16:18:50:41 ...
0
votes
0answers
3 views

Weka class attribute [on hold]

We are trying to run J48 on a classified data set. Our class attribute has two possible values ( 0,1) when running J48 the tree terminates at the very first node and doesnt process any further. ...
0
votes
0answers
11 views

Experimental design - main effects and error bars [on hold]

I've conducted a full factorial design set of experiments, I have 3 factors, with 2 levels for each factor plus a central point to assess the existence of a curvature. I've Measured responses, and ...
0
votes
0answers
12 views

Coherence vs. Magnitude Squared Coherence

Currently I am writing my master thesis. The theory part is about the turbulent wind field generation, where the coherence (not magnitude squared) is used: $$\text{coh}(f) = ...
0
votes
0answers
12 views

Python implementation of Gini impurity [on hold]

I have a 2x2 contingency table and want to compute the gain in Gini impurity (definition extracted from An Empirical Comparison of Selection Measures for Decision-Tree Induction, JOHN MINGERS) I ...
0
votes
0answers
10 views

Mixed effects model for repeated measures to test for factors that are either constant or dynamic within an individual over time

I am dealing with a rather complicated dataset with repeated measures of the same individuals at various time points (samples were collected at different time points and different number of samples ...
0
votes
0answers
19 views

How can I find the variance components from a MCMCglmm output?

I have a dataset in which birds (n=55) were measured twice for five different behaviours. I am now trying to statistically test if individuals are reapeatable/consistent in their behaviours or not. I ...
0
votes
0answers
50 views

How do I solve a under-determined quadratic multi-variate system?

I performed some simulations with known values of input variables $X_1$, $X_2$ and $X_3$, to find output response $Y$. The variables are distributed as following: $$ X_1 = N(\mu_1, \sigma_1) $$ $$ ...
0
votes
0answers
9 views

SPSS Binary logistic regression incoherent result after excluding outliers [duplicate]

I'm using binary logistic regression for my master thesis and after running the regression for a a few specific variables I get the following resutl: Then I created a filter to run the regression ...
0
votes
1answer
14 views

Correlation or dependence between NDVI and pollution data sets

Is there any statistical test or measure to evaluation the degree of correlation or dependence between two sets of data-points ? First set is represented by NDVI values in each pixel and second set is ...
0
votes
0answers
26 views

Establishing consistency

I need to establish the (weak) consistency of an estimator of the mean, $T=a+b\bar{X}$. I tried to apply Chebyshev's inequality, but I couldn't do much because the parameter that subtract in the ...
0
votes
1answer
22 views

Simulation of time series with R [on hold]

I'm new on time series. I'm trying to solve an exercise on the simulation of an ARMA process. The problem is the following: Generate 100 simulations, each with n=60 elements of an ARMA(1,2) process ...
0
votes
0answers
11 views

How is final training stopped? [on hold]

How to know when to stop training the final model, which would be used for test set predictions? There is no validation set anymore... Update: The model is a convolutional neural network trained on ...
0
votes
1answer
28 views

Why do we ignore the main effects when the interactions are not significant in a three-way ANOVA?

I'm following an SPSS guide online, and the procedure is to first test for a three-way interaction. In my case there was no sig. three-way interaction. Then the procedure asks me to consider two-way ...
0
votes
0answers
11 views

Principal component analysis on neural datasets

I have dataset file contains binary bits (0s and 1s) for all attributes Do i still need to apply the Principal component analysis to my dataset to train an autoencoder neural network and when is PCA ...
0
votes
0answers
7 views

cross validation for non parametric clustering methods: dimensionality reduction possible?

I do have about 100 data points gathered during a DoE experiment. The response variable was the settling velocity distribution depending on 10 factors. I analysed the 10 % percentile of the ...
0
votes
0answers
6 views

Package ‘JPSurv’ - Trends in incidence rates [on hold]

Has anyone ever used the JPSurv package in R to look at trends in incidence rates? The original software put out by the NIH - Joinpoint - is not compatible with my computer (a macintosh) and so I ...
0
votes
0answers
16 views

bias of an estimator when the variable is categorical

I want to calculate the bias of the coefficient estimate (beta hat) in Cox regression. But for a categorized variable, for example using quantiles we define a variable with 4 categories, then the ...
0
votes
0answers
22 views

Kernel K nearest neighbours with sparse data

I have a big sparse matrix (around 5 million of lines, 20 000 predictors), and I would like to run a kernelized k-NN on it. However, I don't know how to scale the data properly. So far, I have scaled ...
1
vote
1answer
27 views

Who is the creator or inventor of coefficient of determination (R-squared)?

Who is the creator or inventor of coefficient of determination (R-squared) ? I know that Galton invented correlation.
1
vote
0answers
34 views

Expected time in exponential distribution?

I'm using the exponential distribution to calculate a probability of an event to occur. $\lambda = 0.007$ failures/year. I am to calculate an expected time between 2 failures. My approach was to ...
0
votes
0answers
8 views

Error when using Irlba function in R? [on hold]

irlba(A=X, nu=2, nv=2) Error in matrix(0, m, m_b) : non-numeric matrix extent In addition: Warning messages: 1: In min(m, n) : no non-missing arguments to min; returning Inf 2: In min(n, m) : no ...
6
votes
4answers
562 views

Inverse function of variance

For a given constant number $r$ (e.g. 4), is it possible to find a probability distribution for $X$, so that we have $\mathrm{Var}(X)=r$?
2
votes
0answers
20 views

How does local connection implied in the CNN algorithm

I am trying to understand the process of Convolutional Neural Networks. Basically, I am trying to understand how does the local connection works. The first step of CNN is a convolution layer where ...
2
votes
1answer
24 views

Bars or Boxplot

For my thesis work I need to plot data as dry mass content, Brix, sugar content, etc., of 12 genotypes of onions. My question is, what is convenient to use, barplots or boxplots? Personally I like ...
2
votes
0answers
18 views

Formula of Linear Mixed Effect Model

I would like to perform a regression between two variables and I was wondering if the formula I am using is actually correct. In an experiment, participants had to adjust two parameters of a ...
0
votes
0answers
34 views

How to test whether a random walk have a upward or downward trend

I have applied Kalman Filtering Method to get the estimates for time-varying coefficients (using DLM package in R). The model is like $S_{t} = \alpha_t * A_t + \beta_t * W_t + \gamma_t * A_t * W_t + ...
0
votes
0answers
7 views

Jackknife validation in linear regression for model selection

I have built a linear regression model linking Temperature (T) to metal expansion (E) rates using n = 30 data points which I collected from measurement: T = aE + b where a and b are the regression ...
0
votes
0answers
20 views

Understanding contradictory results between screeplot (EFA) and CFA

I got some contradictory results. How can one explain that a screeplot and EFA clearly suggest a one-factor solution whereas several model fit parameters of CFA are in favor of a five-factor solution? ...
-1
votes
2answers
32 views

non-invertible weakly stationary AR processes

ok, so I know (for example) that $X_t = 2X_{t-1} + e_t$ for iid $e_t$ is stationary. But how do I go about proving the condition $Cov(X_r, X_s) = Cov(X_{r+t}, X_{s+t})$ for all integer $r, s, t$ ...
0
votes
0answers
10 views

Generate random variables correlated with dependent variables without Cholesky in R

I want to generate random variables correlated with dependent variables in R. I dont want to create multicollinearity between random variables. I try to create a function: Some of these random ...
1
vote
1answer
19 views

R - Multivariate K-nearest neighbor outlier detection

I'm trying to implement the algorithm K-nearest neighbor to detect outlier from a multivariate dataset. I don't know how to do it. Could you provide me some example?
1
vote
1answer
64 views

Using linear regression for count data - will this introduce bias?

Say I am fitting a model to Poisson count data, but I am only interested in estimating the mean of the count variable. I understand a ordinary linear regression is a good approximation when the ...
0
votes
0answers
13 views

Multiple DVs - MANOVA or multiple ANOVAs?

My project has multiple continuous DVs and a single categorical IV. When I run ANOVA for each of my DVs the means and SDs are different to when I use MANOVA. Why is that? Which one is correct? Which ...
1
vote
1answer
19 views

How to check if the correlation between two continous variables is influenced by a categorical factor?

I have a data frame (df) where I see correlation between two continuous variables (c1 and c2). I need to know whether the observed correlation between the two variables differs between groups, which ...
0
votes
1answer
10 views

Frequent sequence mining SPADE with sequences of binary codes [on hold]

I have sequences that look like: A-A-A-B-B-A-A-A, with only As and Bs. I ran SPADE on it to find frequent patterns, but only got 3 frequent sequences: A, or B, or A-B. Why are patterns like B-A, or ...
0
votes
2answers
19 views

Detrending a series that moves about zero

How can I avoid my de-trending model from blowing up? Do I need an additive model rather than a multiplicative model? If so, is there anything I would consequently need to take into account? Can I ...
0
votes
0answers
10 views

How to make Likert Charts in SPSS [on hold]

Hello I am a beginner in doing Likert charts and I was wondering if there is any video on internet that can guide me on doing Likert charts using SPSS. My questionnaire have many questions that used ...
0
votes
1answer
21 views

using an odds ratio to predict prevalence/incidence in different populations

Say you have an odds ratio which represents the ratio of outcome given a particular exposure to the outcome given no particular exposure. How can you use this ratio to make an estimate about the ...
0
votes
1answer
12 views

Cross-validating the tbats/bats function in forecast

Is there a way to cross validate the tbats/bats function in the forecast package in R? I have been trying to get CV weighted parameters which then I can pass to a function for revised estimates. ...
0
votes
0answers
5 views

How do I find the expected coverage of the domain after K ranges are selected?

If I have a domain of 0 to 1000000 (but I doubt the domain matters, just that the ranges can overlap). I then chose 1% ranges uniformly, allowing for ranges to overlap completely or partially. How do ...
0
votes
0answers
19 views

Is it essential to use a correlation matrix when the scales of variables are different in PCA?

I am doing PCA on two variables which have completely different units and scale. More exactly, I am doing multivariate-EOF analysis as the columns of the data matrix represent 12 grid points of the ...

15 30 50 per page