Questions tagged [explanatory-models]

Models created to explain a response (as opposed to simply predict it). This is generally understood to imply models of causal processes, or to test hypothesized causal relationships.

Filter by
Sorted by
Tagged with
1 vote
0 answers
25 views

Appropriateness of using SHAP values to evaluate a model

I am a deep learning researcher that would like to use SHAP values to assess the relative importance of input features on the model's final score. Colleagues of mine have taken issue with the method ...
user avatar
  • 21
1 vote
0 answers
31 views

interpret lime results

What is the meaning of the values of the "blue" features in the lime output? I understand that they influence the lime black-box model to classify an observation as 0 label, but what is the ...
user avatar
  • 67
0 votes
1 answer
33 views

Use of simulation in explanatory versus predictive models

I understand the important distinction between explanatory vs. predictive models. However, I often read that simulations is an important tool in risk modelling and risk analysis. How are simulations ...
user avatar
0 votes
0 answers
19 views

How Expected Prediction Error and Parameter estimation error are related each other?

I grasp that the ultimate goal of regression is essential for proper optimization: If your goal is prediction, you must minimize the prediction error. Otherwise, if your goal is explanation, you need ...
user avatar
  • 1
1 vote
1 answer
63 views

How explainable is Linear Discriminant Analysis?

In a survey paper on the interpretability of various machine learning algorithms, I didn't find Linear Discriminant Analysis (LDA). I wonder how explainable is LDA to audiences not familiar with ...
user avatar
0 votes
0 answers
41 views

(Explanatory) Modelling with PMG and MG estimators

Hello kind people of Stackexchange. I happen to be kind of stuck and in need of help. See, I want to make an ARDL model with PMG and/or MG estimator (Pesaran, 1999) (okay, in fact it may be the CCE ...
user avatar
  • 21
1 vote
1 answer
28 views

Clarification the function of coefficient of Post variable in DID for two-year two-group

When reading a paper about DiD, I saw a sentence describing the function of the $Post_t$ variable in DiD for two groups and two periods: In a similar manner, the coefficient on $P_t$ captures the ...
user avatar
0 votes
0 answers
20 views

Metrics for Explanatory Power of Machine Learning Classification Models

I am interested in finding metrics that assess the explanatory power of machine learning models involving binary classification. I know that for logistic regression, McFadden pseudo R squared, AIC, ...
user avatar
0 votes
1 answer
53 views

Explainable neural network through supervision of penultimate layer

Generally, most deep learning (DL) models are considered opaque black boxes, and post-hoc explanations may not be satisfactory to users, especially for use cases in the legal or clinical world where ...
user avatar
0 votes
0 answers
50 views

How to explain features' importance after PCA?

I had a dataset that was terribly haunted by colinearity. I carried out PCA on it to remove the colinearity. Let's assume I am going to build a linear regression model, which is easily explainable, on ...
user avatar
0 votes
2 answers
2k views

Why Feature Selection with sklearn.feature_selection.SequentialFeatureSelector is a preprocessing task?

I am facing a feature selection problem. Because I am building an Explanatory Regression Model I decided to follow a Forward Sequential Feature Selection. Moreover I wanted to implement ...
user avatar
0 votes
1 answer
56 views

X on Y Linear regression

Basically in a research project I am looking at the linear regression between my independent variable: Government Stringency Index, and dependent real GDP growth. One area I investigate assumes if ...
user avatar
2 votes
0 answers
49 views

Assessing potential multicollinearity for various model types meant for causal inference [duplicate]

I was wondering if, in the case where I use the same set of predictors for different kind of models (e.g. ANOVA, Poisson GLM, logistic regression...), VIF values for each predictor would be similar ...
user avatar
  • 383
4 votes
2 answers
73 views

Is there a way to increase the number of predictor in a logistic regression when sample size is small?

One of the dependent variables in my dataset is a binomial variable indicating whether an invasive species has been eradicated (with $n_{1}$ = 23) or not (with $n_{0}$ = 75) after the use of a certain ...
user avatar
  • 383
3 votes
1 answer
196 views

Is a beta regression appropriate for a skewed bounded continuous dependent variable when sample size is small?

My data contain a bounded continuous variable (score between 0 and 10) representing the efficacy of a given method to control an invasive species. As there are more high scores than low ones, the ...
user avatar
  • 383
-1 votes
2 answers
21 views

Another term for "real time decisioning"

There is another term for "real time decisioning," (RTD) and I need it to locate a paper (by searching). I've been beating my head over this all day. I'm thinking it's "[something] ...
user avatar
  • 1
2 votes
0 answers
24 views

Redundancy analysis (RDA) to identify the relationship between water quality and land use

I have read some papers in where the authors had performed redundancy analysis (RDA) to identify the relationship between water quality and land use. However, I am confused as to how they are setting ...
user avatar
  • 143
3 votes
1 answer
405 views

Why do DeconvNet use ReLU in the backward pass?

Why does DeconvNet (Zeiler, 2014) use ReLU in the backward pass (after unpooling)? Are not the feature maps values already positive due to the ReLU in the forward pass? So, why do the authors apply ...
user avatar
2 votes
0 answers
70 views

Should I use cross-validation to build an explanatory model?

I know it is recommended to do so for a predictive model, but if I am trying to study a potentially causal relationship and adjusting for covariates to be able to isolate that relationship should I ...
user avatar
  • 21
5 votes
2 answers
700 views

Reasons that LIME and SHAP might not agree with intuition

I'm leveraging the Python packages lime and shap to explain single (test-set) predictions that a basic, trained model is making on new, tabular data. WLOG, the explanations generated by both methods ...
user avatar
1 vote
3 answers
84 views

Is a predictive model the best way to generalize about my current data?

Suppose I have a customer dataset. My independent variables are various types of customer attributes (age, where they live, gender, price of the item they are potentially buying) and my dependent ...
user avatar
  • 117
18 votes
4 answers
772 views

On George Box, Galit Shmueli and the scientific method?

(This question might seem like it is better suited for the Philosophy SE. I am hoping that statisticians can clarify my misconceptions about Box's and Shmueli's statements, hence I am posting it here)....
user avatar
  • 10.7k
0 votes
0 answers
629 views

Application of GAM on large dataset

I was suggested that my questions were too broad. As I commented below, I have nearly a million data points and perhaps a hundred variables. This may be a very basic modeling question: I am curious to ...
user avatar
1 vote
1 answer
90 views

What are the most important/seminal/popular methods in the field of interpretable ML?

As I've been trying to get a grasp on the field of interpretable ML I have encountered many interesting papers and methods, but I lack the perspective to determine their importance to the field as a ...
user avatar
0 votes
1 answer
146 views

In scale adaptation which should be done first? CFA or Reliability Analysis

I am adapting questionnaire cross-cultural. Questinonnaire has 11 item and two factor. I want to confirm the structure 1 item's corrected item total correlation is <.20. Should this item be ...
user avatar
  • 1
0 votes
1 answer
46 views

Is this an appropriate way of modeling the scores in a round-robin sports league?

I want to model the outcome of matches in a round-robin sports league based on which home team is playing which away team across several seasons. Let's assume a league with four teams ...
user avatar
  • 123
6 votes
2 answers
185 views

Does cross validation say anything about parsimony?

Suppose I had a set of models that all attempt to explain some phenomena. According to a sensible—and appropriately cross-validated—performance metric, all of the models perform comparably well. The ...
user avatar
  • 19.3k
3 votes
2 answers
159 views

Standardization and explanatory variables of different domains in Multiple Regression

There's many questions on related topics but I have been unable to find one that precisely answers my question. Let's say I'm performing a regression on multiple predictor variables $x_1...x_n$ for ...
user avatar
  • 1,515
0 votes
0 answers
61 views

How does LIME compares with Mutual Information?

So, I was wondering how LIME's linear model approach compares with other explanation metrics, in special, with Mutual Information? For those unfamiliar with how LIME works: Choose the instance you ...
user avatar
6 votes
3 answers
229 views

General approaches and techniques for developing good explanatory models for nonlinear data

Various recent efforts of mine on modelling some data through logistic regression have been... not successful. While there is still more data to look at, I've been wanting to explore nonlinear ...
user avatar
  • 1,515
2 votes
1 answer
87 views

Quantifying explanatory potential

Suppose I have a random variable $T_j \sim Bernoulli(p_j)$ where $logit(p_j) = \theta x_j + \epsilon_j$ and where $\epsilon_j \sim \mathcal{N}(0,1)$. Suppose further that $\theta = 0.018$ and that I ...
user avatar
20 votes
2 answers
508 views

Minimizing bias in explanatory modeling, why? (Galit Shmueli's "To Explain or to Predict")

This question references Galit Shmueli's paper "To Explain or to Predict". Specifically, in section 1.5, "Explaining and Prediction are Different", Professor Shmueli writes: In explanatory ...
user avatar
8 votes
4 answers
18k views

Is MSE decreasing with increasing number of explanatory variables?

I am wondering, if there is a negative correlation between Mean Squared Error \begin{equation} MSE = \frac{1}{n} \sum (\hat{Y}_i - Y_i)^2 \end{equation} and the number of explanatory variables. ...
user avatar
2 votes
0 answers
2k views

How exactly does ridge regression helps in the case of multicollinearity?

I understand the reasoning behind ridge regression: we include some bias in the model in order to reduce the variance of the regression coefficients. My question is, why would we want to do that? ...
user avatar
2 votes
4 answers
423 views

Purpose of leave-one-out cross-validation in descriptive modelling

I refer you to Breiman's paper Statistical Modeling - A Tale of Two Cultures where he illustrated some examples of descriptive modelling. Under section 11.1, 100 runs of regression were performed, ...
user avatar
  • 695
7 votes
1 answer
123 views

Is inference based on full (global) regression model appropriate?

Is inference based on a full model appropriate, and if so, in which circumstances? Suppose you are interested in the potential relationship between a response variable and several candidate predictor ...
user avatar
  • 800
14 votes
1 answer
838 views

Should I use unpenalized logistic regression, lasso or ridge for explanatory modelling?

When using logistic regression for predictive modelling, the choice between 'standard' logistic regression vs ridge vs LASSO versions of logistic regression seems relatively straightforward - just ...
user avatar
  • 1,075
0 votes
1 answer
367 views

How to test predictive ability of an explanatory model?

As a part of my research I create an explanatory negative binomial regression model. Now, I want to show this model can also have predictability power. I don't want to compare my model with other ...
user avatar
4 votes
3 answers
14k views

Predictive power vs. explanatory power of statistical models

Is it possible for a statistical model to have explanatory power but no predictive power?
user avatar
  • 41
1 vote
0 answers
62 views

Assess a model building technique

I am confused about a certain model building technique that seems to exist, at least in practice (I am not sure whether it has its place in textbooks). Question 1: I wonder under what conditions or ...
user avatar
9 votes
1 answer
630 views

LASSO for explanatory models: shrinked parameters or not?

I'm conducting an analysis where the primary goal is to understand the data. The dataset is large enough for cross-validation (10k), and predictors include both continuous and dummy variables, and the ...
user avatar
1 vote
0 answers
113 views

A bunch of different types of variables (their combination also important) explaining one variable - which method?

I have a dependent variable - how much land does a household cultivate out of total in their possession. The answers are categorized in 3 different groups (1 - 70% - 100%, 2 - 40 - 70%, 3 - less than ...
user avatar
4 votes
1 answer
1k views

Prediction vs. Explanation and its Effect on Statistical Methods [duplicate]

In layman's terms, what is the difference between predicting and explaining in statistics? I was looking for the differences between AIC and BIC and found this post with an answer stating: My quick ...
user avatar
  • 701
1 vote
1 answer
285 views

Predictive modeling techniques for in-sample rather than out-of-sample prediction?

Is it appropriate to apply predictive modeling variable selection and shrinkage techniques (for example, ridge regression or lasso) for in-sample prediction rather than out-of-sample prediction? ...
user avatar
  • 4,504