A statistical model is a formalization of relationships between variables in the form of mathematical equations. The model is statistical as the variables are not deterministically but stochastically related.

learn more… | top users | synonyms

1
vote
0answers
8 views

Transforming / Evaluating (Probably) Normal Data from Sample with Linear Probability Distribution

I have a sample of data for which the count of a given value increases roughly in proportion to the actual value. For example, the value 22 will occur around 22 times and the value 30 will occur ...
0
votes
0answers
17 views

Linear mixed model when interest is ONLY in population

Lets say we have 1 to 6 repeted measures over time for some (many) subjects, and the outcome is values between 0 and 23 (so we assume normal distribution). For a normal distribution the fixed effects ...
0
votes
0answers
24 views

How to adjust for a continious variable when the value 0 is distinctly different from the others? [duplicate]

Lets say I want to regress a variable on a covariate which has distinct value zero but the all other values are following some smooth function? It could be e.g. Years of beeing a mom (once you are a ...
8
votes
1answer
82 views

Would there be a model selection problem if we had access to an oracle that gave us the exact generalization error?

Let $\mathcal{E(h)}$ a function that given some hypothesis $h$ returns the generalization error for that fixed $h$. I was reading some notes about model selection and generalization error and it ...
0
votes
2answers
42 views

Ensemble of models with different feature spaces

BACKGROUND I have data in which the dependent variable is binary with a highly-skewed distribution: <1% records are 1 (doers), >99% records are 0 (non-doers). I'm using logistic regression to ...
2
votes
2answers
58 views

Same dataset analysed with four different linear models

I've analysed the same dataset (diamonds from ggplot2) in R with four linear models. Each model has a different error structure. ...
0
votes
0answers
27 views

What would you use as your default model for comparison?

I have a question. Say you are given with like 1500 instances as your dataset, in which instance will be categorized into 1 of the 3 classes. You're supposed to come up with the best model from the ...
0
votes
0answers
43 views

Caret with rxDForest as custom model

I'm trying to use the train function from the caret package to tune the parameters of the rxDForest from RevoScaleR package (I ...
0
votes
0answers
5 views

Statistical models on Birth interval data

Birth interval means time between successive births. I have a birth interval data from DHS survey. If we want to see the effect of familial and socioeconomic co-variates on birth interval. How can we ...
1
vote
0answers
23 views

Can conditional distributions be considered as a parameterized family of distributions?

Given two real-valued random variables $X$ and $Y$, are the conditional distributions of $Y$ given $X$ taking different values, $ \{ p(y|x), \forall x\}$, considered as a parameterized family of ...
0
votes
0answers
45 views

Interpreting coefficient, marginal effect from Linear Probability Model

I am regressing part time as a binary dependent variable (0 who dont work part time and 1 people work part time) with different parameter listed below partime – variable=1 if employee works part ...
1
vote
0answers
23 views

How to determine appropriate number of features and also which features to select?

So I have a dataset which I am using K fold cross validation on to select the number of features and which features should be selected. As I understand it, I would set the number of features to be ...
0
votes
0answers
38 views

over estimating ARIMA model

When an ARIMA model over estimates, i.e if the forecast from the model is always higher than the actual value, what's the cause? The model i used was SARIMA (1,1,0)x(1,0,0). Below is the time series ...
1
vote
0answers
35 views

Sequence of Bernouilli trials with diminishing returns

Say I have a sequence of yes-no trials where the probability of winning decays over time as the supply of wins is gradually exhausted. Assume that for every trial the probability of replacement is not ...
1
vote
0answers
30 views

Choosing an appropriate model - What does this phrase mean?

I have a question where I am given a short story: An investigator wishes to evaluate the effect of medicine administered for colds on reaction time. Data represent the reaction times (in ...
0
votes
0answers
20 views

statistical distribution that would generate combination of numbers without repitition

I have set of 'n' numbers in an array. Is there a statistical distribution that would generate randomly picked 'n' numbers from the same array such that there are no duplication. For example, if I ...
0
votes
0answers
17 views

How to estimate False Discovery Rate from p-value distribution?

I have learned many models and I calculated p-values for the cross-validation errors. I want to select significant models based on the false discovery rate (FDR). How can I estimate the FDR from ...
0
votes
0answers
20 views

likelihood of a model

the likelihood of a model is defined as the probability of data given model: Likelihood(Model) = p(DATApoints | Model) which is equivalent to the product of all p(datapoint | Model) for each ...
1
vote
1answer
127 views

AIC BIC Mallows Cp Cross Validation Model Selection

If you have several linear models, say model1, model2 and model3, how would you cross-validate it to pick the best model? (In R) I'm wondering this because my AIC and BIC for each model are not ...
1
vote
0answers
44 views

Negative binomial distribution mixture model with R

I have two data vectors of observed count data: $A$ and $B$, where count $A_n$ and $B_n$ refer to the same observation point. $A$ is assumed to follow a negative binomial distribution. $B$ is assumed ...
2
votes
0answers
48 views

Binary Class Distribution Effects on Probability Scores - (gbm) Boosted Tree Regression Models

Any help would be greatly appreciated. Problem: I need help to better understand the probability scores that come from the result of a decision tree model. Specifically, I'm using the gbm package ...
2
votes
1answer
26 views

Is including the main effect of the covariate enough?

Is simply including a covariate in a model (ANCOVA) enough for variance in the DV due to it to be factored away? Or do the higher order interactions of the covariate with the other factors of the ...
1
vote
1answer
24 views

Change the classifier decision by using the probability estimates

I have a stream of documents composed of $1$ to $n$ pages. The objective is to segment the stream of documents. Every first and last page of a document is classified as either the beginning $b$ or ...
0
votes
0answers
17 views

Linear Mixed Model Construction Suggestion

I'm currently working on some data that requires me to use linear mixed model. Explanation of Data Experiment of Drugs on Mouse I have 3 drugs and 1 control Unbalanced number of mouse for each ...
1
vote
0answers
13 views

Analysis of model input resolution

I am using a fine resolution soil dataset to run a climate model, and want to explore the impact of aggregating the soil properties at different resolutions on model output. E.g. how do the results ...
0
votes
1answer
32 views

Collapse dimension when comparing model coefficients

I have a dataset that includes 2 dimensions (A, B), each of which has the same two categories (1,2). I fit a model to the data in each of the 4 conditions and get 4 coefficients for a parameter of ...
5
votes
0answers
36 views

Geometric interpretation of generalized linear model

For linear model $y=x*\beta+e$, we can have a nice geometric interpretation of estimated model via OLS: $\hat{y}=x*\hat{\beta}+\hat{e}$. $\hat{y}$ is the projection of y onto the space spanned by x ...
3
votes
1answer
64 views

Do fewer support vectors imply a simpler model?

I am applying $\epsilon$- and $\nu$-regression to sample data, and I discovered I had different results in terms of the count of support vectors. When I have fewer support vectors, does it mean that ...
0
votes
0answers
26 views

AIC, BIC: Multiple dependent variables

I have multiple candidate models, and each needs to be scored on multiple "metrics", where each metric is probability distribution. Essentially the metrics are like dependent variables. The question ...
0
votes
0answers
22 views

how to calculate baseline hazard rate for proportional hazard model with time varying covariates

I fitted a proportional hazard model with time varying covariates. However SAS does not produce the baseline hazard rates for the model with time varying covariates. I am wondering if there is an easy ...
3
votes
1answer
54 views

Determining if a function is additive

I have data collected on a grid of values $f(x,y)$. I have a hypothesis that the data is "additively separable", i.e. $$ f(x,y) = h_1(x) + h_2(y) + g(x,y) $$ where $g(x,y)$ is small and only ...
0
votes
0answers
19 views

Can you create a composite score (e.g. index) out of scales that form a mediated model?

I have a mediated model of decision making with three elements - values, attitudes and intentions. The relationship between values and intentions is mediated by attitudes. Each element is measured ...
0
votes
1answer
126 views

How to validate a Multinomial Logit and Probit Model fit?

I would like to know how do you determine the performance of your models. That is, if you fit a multinomial logit or probit model for un-ordered discrete choice. What do you use to evaluate whether ...
4
votes
1answer
82 views

Predict best action sequence for dataset of debt observations

I need some suggestions of general predicting practice in the following case: I have a dataset of debt observations; there are ~8 variables defining each debtor situation (debt details and person ...
2
votes
2answers
80 views

ACF of the errors

For revision, I am working out a multiplicative model for sales data, and then conducting a simple error analysis (actual sales - forecasted). I understand the process but in my lecturer's mark scheme ...
2
votes
0answers
35 views

AIC doesn't agree with model checking [duplicate]

I have two glm, one with a gaussian distribution and identity link and one with gamma family and log link. The predictors are the same, the only thing that change is the response that in the gaussian ...
0
votes
0answers
27 views

How to measure the accuracy of multinomial target?

I have trained a model that predicts a multinomial output. Would anyone know how the accuracy can be measured. The target takes on the following values: "Yellow","Red,"Green","Blue". I know that ...
2
votes
0answers
108 views

How to make predictions from a logistic regression by hand

Thanks for looking at this, I've been tearing my hair out for a day or so now. I have done a multiple variable logistic regression in R, and obtained my coefficients. I am able to make predictions ...
1
vote
1answer
37 views

which performance metrics to classify model

I wonder between two performance metrics for classification models: accuracy and area under ROC curve (AUC), which one is to be preferred in which conditions? examples appreciated
1
vote
0answers
24 views

Single-class vs two-classes hypothesis testing

Consider a Gaussian mixture model of unknown parameters, having just one or two components. I would like to design a statistical test to decide the number of classes ($H_0$: single component model vs. ...
1
vote
0answers
55 views

Am I using the correct model?

I'm new to statistics. What type of model should be used to predict a basketball team's winning percentage? I am currently using a multiple linear model, but the residual vs. predicted plot shows ...
3
votes
0answers
70 views

Which technique to build a model returning a vector of values (in R)

In my current project I need to build a model returning a vector of actions for each observation. I need a suggestion which statistical technique is used in general in such cases. In a project, I ...
2
votes
1answer
102 views

If ϵ is uniformly distributed, then a linear probability model is appropriate? Can I find any Literature?

A latent variable model involving a binomial observed variable $Y$ can be constructed such that $Y$ is related to the latent variable $Y^*$ via $ Y = \begin{cases} 0, & \mbox{if ...
5
votes
3answers
134 views

Why is KNN not “model-based”?

ESL chapter 2.4 seems to classify linear regression as "model-based", because it assumes $f(x) \approx x\cdot\beta$, whereas no similar approximation is stated for k-nearest neighbors. But aren't both ...
0
votes
1answer
44 views

Extract Information from Traded volumes

Assume that we want to extract information from a constantly updating table like stock trading volume, where a banker can buy or sell at specific price. The table has the following 4 columns: price, ...
3
votes
1answer
73 views

Use of expectation propagation for model inference

I have a joint probability distribution as given in the figure: In this figure, variables in circles are random variables and variables in squares are constants. So, I can write the joint ...
0
votes
3answers
79 views

Statistic model for removing bad data

Let's say I have a (ads live days, revenue) data set. The data set shows how much revenues each ads generates during the days it is live. ads1 generates 100 dollars during the 5 days when it is ...
1
vote
0answers
20 views

Collectively evaluate a number of normal distributions [duplicate]

I build a few models, each model will produce a normal distribution for the value of a future event. For example, model M1 will produce a normal distribution $n(30, 5^2)$, and the value of the future ...
5
votes
1answer
50 views

Model selection criterion produces non-normal residuals

I was wondering if I can use Akaike or Schwarz criterion even when the residuals that I get from the model when I run the regression are not normal. Is there any normality assumption with these ...
3
votes
1answer
99 views

Mixed Effects Model with Nesting

I have data collected from an experiment organized as follows: Two sites, each with 30 trees. 15 are treated, 15 are control at each site. From each tree, we sample three pieces of the stem, and ...