Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [model-interpretation]

The tag has no usage guidance.

0
votes
0answers
13 views

Interpreting $p$-values and Chi square when variables are added to a model in maximum likelihood estimation

I'm running a maximum likelihood estimation (probit) and I'm experimenting with adding additional variables, walking the bias-collinearity tightrope. Please could someone explain to me intuitively ...
0
votes
0answers
5 views

LMM estimates and effect plots don't match

I have a cubic model with interaction terms between continuous predictors, of the form: ...
0
votes
0answers
11 views

How to quantify the effect of the inputs and the effect of the labels on our regression model?

Let us imagine in a regression setting where our responses are D dimensional vectors, and so are our inputs. Let's say I train a regression model (e.g, Neural network). When I update my weights on a ...
0
votes
1answer
22 views
0
votes
0answers
20 views

Biological Interpretation of a Cox Model

I'm interested in a continuous variable (let's say blood pressure) and it's role in the disease progression (let's say heart attack). Now, let's say the literature has established already body fat and ...
0
votes
0answers
7 views

Does a feature map order prior exist in deep learning?

Does a feature map order prior exist in deep learning ? Example: making sure that feature map i has the same meaning from one training to another (with or without exactly the same dataset)? I'm ...
2
votes
0answers
79 views

Interpreting interaction term on highly correlated variables

Somebody at a meeting today made the following comment about a Marketing Mix Model (Linear Regression) we run every year. We should account for the high collinearity of the two Marketing Variables (...
0
votes
1answer
21 views

Percent Censoring in A Cox Regression

How much right-censoring in a PH model is acceptable? For example, my summary of the number of event and censored values is Total: 450 Events: 200 Censored: 250 Percent Censored: 55.6% PH ...
0
votes
0answers
32 views

Arima model question

I have some basic Arima questions.I built the following two Arima models in R (output below) and the R output each time mentions something about the mean being non-zero or zero. ...
0
votes
0answers
21 views

Hurdle models and word count

Question: why negative binominal part of hurdle model does not provide coefficients for intercept and word count? I have counted positive emotional words (Y) in some conversations that have a ...
1
vote
1answer
46 views

Model Quasipoisson interpretation and validation

I am currently doing my Master thesis with evaluating my results in R. I am stuck on my analysis of my glm with quasipoisson. I am analysing influencing variables on the dormouse abundance in 2 types ...
1
vote
1answer
26 views

Difference between averaging and ignoring the partial dependencies?

This question sparks from model interpretation/visualization. To graph the dependency of a function with >2 arguments, one often needs to ignore or average out some arguments. Problem set Hastie, ...
0
votes
1answer
81 views

How to interpret the results from a binary model done in R? [duplicate]

This is a binary model that my colleagues and I developed that regards a connection between the distance (dependent variable) and overall satisfaction with public transport (covariate). I don't know ...
0
votes
0answers
24 views

Why use absolute value of the gradient for saliency maps?

Simonyan et al. 2014 approximate a neural network locally by a linear function and take the weights of this approximation as a measure for support of a specific class. Can someone explain to me why ...
1
vote
1answer
76 views

RMSEA confidence interval interpretation

Using Amos for a multi-group path analysis, my model of choice, the "measurement weights" model has an RMSEA value of .033 a LO value of .000, a Hi value of .065, and PCLOSE value of .784. Is it an ...
0
votes
0answers
53 views

The average change interpretation when the dependent variable is in logarithm

Suppose the regression model is $\text{wage} = \beta_1x_1 + u$. The conditional expectation function is $\operatorname{E}[\text{wage} \mid x_1] = \beta_1x_1$. Then, we interpet $\beta_1$ as the change ...
1
vote
0answers
157 views

What are the differences between LIME and SHAP as model interpretation techniques?

Model interpretation has been an important area of study nowadays and because of that some different techniques have risen to help on this task. Maybe the two most famous ones are LIME and SHAP. How ...
0
votes
1answer
27 views

multiple regression - interpreting the plots of the model

I'm doing a multiple regression analysis with up to 9403 observations without any prior hypothesis. The AIC and BIC variable selection both generated the same ideal model. After I ...
11
votes
1answer
1k views

Comparison between SHAP (Shapley Additive Explanation) and LIME (Local Interpretable Model-Agnostic Explanations)

I am reading up about two popular post hoc model interpretability techniques: LIME and SHAP I am having trouble understanding the key difference in these two techniques. To quote Scott Lundberg, ...
0
votes
0answers
42 views

A question concerning the SHAP paper (A Unified Approach to Interpreting Model Predictions)

On page 6 of this paper, it is mentioned that the LIME model falls in the case of Equation 12 in that paper. My question is, in what sense? More specifically, Equation 12 approximates $f(h_x(z'))$ ...
1
vote
1answer
217 views

Why the discrepancy between predict.xgb.Booster & xgboostexplainer prediction contributions?

One way to explain individual predictions of an xgb classifier is to calculate contributions of each feature. To my knowledge there are two packages in R that can do this for you automatically. In the ...
3
votes
1answer
185 views

Random Forests interpretability [duplicate]

I have been using the sklearn RandomForestClassifier to solve a binary classification problem. For a particular sample prediction, I would like to be able to know ...
1
vote
1answer
119 views

How to interpret a seasonal ARIMA model?

I have an auto.arima model output with ARIMA(000)(110)[4] with sigma^2=0.005, so I assume the model fits well the data. But I'm trying to understand the model itself... If I did all the maths ...
0
votes
0answers
152 views

Interpreting Accumulated Impulse Response graphs for SVAR Models

I am doing a study similar to this one using Eviews 10 https://researchportal.port.ac.uk/portal/files/189238/Economic_Modelling_GF.pdf except with updated data just for the UK. I have produced ...
0
votes
1answer
47 views

Interpreting Regression Coefficient

I am using panel data and I have a very simple regression equation of the following form: $y_{st} = \alpha_s + \beta_t + \gamma F_{st} + \epsilon$ Where, $y_{st}$ is my dependent variable (...
0
votes
0answers
17 views

Second stage in TS2SLS

I am working with a Two Stages Two Steps Least Squares (TS2SLS) in a Diff-in-Diff context, with cross-section data. I have a large sample where I observe the binary outcome Y as well as the ...
0
votes
2answers
346 views

How can I interpret intercept when my dependent variable is in log form?

My model consists of both log independent and dependent variables, and percentage share, as well as number of people and etc. My dependent variable is log(GPD per capita in USD), my statistic package ...
0
votes
0answers
89 views

Help interpreting an interaction term from R output of linear regression

So I am familiar with the concept interaction. However in most instances that I have come across, a significant interaction term is in the same direction as the 2 individual terms. However I have come ...
0
votes
0answers
42 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 ...
4
votes
2answers
96 views

Finding most similar training samples for a given ML model output

Let's say you have trained a regression model. Now, in production, you get a new input, and you want to identify K samples in the training set that are the most similar to the new input as far as the ...
0
votes
0answers
63 views

How to interpret a VAR model without significant coefficients?

I am trying to investigate the relationship between some Google Trends Data and Stock Prices. I performed the augmented ADF Test and KPSS test to make sure that both time series are integrated of the ...
1
vote
2answers
59 views

Understand my linear model

How can I explain that a value, first positive correlated with the target, has a negative influence on the target in my model? I have built a linear model in R. Had 30 variables, and 60 observations....
0
votes
1answer
31 views

How to interpret statistical insignificance of explanatory variables that may be indirectly affecting Y?

I have modeled Y (binary variable) on a set of explanatory variables, X, using logistic regression. A number of x's, say 10 of them, are found to be significantly affecting Y (based on LR test). Then ...
0
votes
1answer
85 views

Regression with percentage dependent, log and linear independent

I have read a lot of questions and answers but I am still confused for my case. My dependent variable (employment growth) is measured : $$ \frac{((number\ of\ employees \ in \ year\ t) - (number \ of ...
2
votes
0answers
36 views

To impute or not - community consensus for reporting accuracy of an imputed model

I have a model generated using an imputed data set with imputation accuracy of 75%. If the model using imputed data has an accuracy of 80% What would be the community consensus to report the ...