Questions tagged [predictive-models]

Predictive models are statistical models whose primary purpose is to predict other observations of a system optimally, as opposed to models whose purpose is to test a particular hypothesis or explain a phenomenon mechanistically. As such, predictive models place less emphasis on interpretability and more emphasis on performance.

Filter by
Sorted by
Tagged with
1
vote
1answer
27 views

Normalise different thresholds for binary prediction

I'm working in a module that outputs the risk of an event happening i.e. risk of a crime happening depending on the district of the city. What I've done is to calculate for each district a binary ...
0
votes
0answers
27 views

Including interaction term without main term with possible aliasing

I'd like to model an interaction term between a continuous variable and categorical variable, while accounting for possible aliasing in the variables. I was wondering what the best way to do this was. ...
1
vote
0answers
18 views

Predicting outputs for new samples in Bayesian linear regression

This is my first question on this forum. I just got started with Bayesian statistics. While I do understand the motivations behind Bayesian methods, I am a little unclear on what the predictions even ...
0
votes
1answer
27 views

Difference Between Sensitivity Rate & Hit Rate

I've been told by my professor that hit rate is more important when it comes to real-life deployment but sensitivity and other accuracy rates need to be satisfactory also. However, my other professor ...
0
votes
0answers
7 views

improving link prediction model on knowledge graph so that it does well on anti-symmetric relation type facts

I was trying to look at the task of link prediciton on the WNR11 dataset. I looked at the TuckER model and found that the TuckER model can do very well on the facts involving symmetric relation types. ...
1
vote
0answers
21 views

Q: AUC of each subgroup is smaller than overall AUC

I have a validation data set of 29242 patients, with known labels/health outcomes and predictions that were generated by some model. 28626 patients are negative and 616 are positive The overall AUC is ...
0
votes
0answers
9 views

Correlation between classes within categorical feature in a ML model

I have a dataset containing numerical and categorical features. Let's say I have one categorical feature C with classes c_0 ... c_n. I build a model that has some ...
0
votes
0answers
9 views

Can you have a concordance statistic (discrimination) when predicting a continuous outcome?

I'm writing a proposal for a prediction model predicting BP (continuous outcome, predicting trend over time). For assessing model performance, I'm seeing discrimination and calibration as the most ...
1
vote
0answers
18 views

What type of algorithm should I use to analyse questionnaires answers?

Let's say I have a questionnaire, part of it are multiple-choice answers. The person answering may answer (a), (b) or (c) for the first question, then (a), (b), (c), (d), (e) or (f) for the second ...
0
votes
1answer
17 views

Combine a probability from a prediction model and the disease prevalence?

I created a prediction model using a multiple logistic regression on a non-representative sample of the target population. With this model, I can get a probability for a new patient of belonging to ...
0
votes
0answers
25 views

Event-based time-series analysis

I'm trying to find patterns in a game event that happens randomly over time. The data looks like this: ...
0
votes
0answers
26 views

What is the best way to present the following predictive regression relationship?

If I have a predictive regression with a single regressor of the form \begin{equation} y_t=\beta x_{t-1}+\varepsilon_t \end{equation} where \begin{equation} x_t=\rho x_{t-1}+u_t \end{equation} Then I ...
0
votes
1answer
87 views

How do I interpret mean absolute error (MAE) or mean absolute percentage error (MAPE) in layman words?

For example, I am predicting a score that can have value from 0 to 100. Lets assume MAPE = 10...
0
votes
0answers
28 views

How do I interpret RMSE in layman words? [duplicate]

For example, I am predicting a score that can have value from 0 to 100. The RMSE = 10. How ...
0
votes
0answers
10 views

How do I approach this problem?

Let's say I have a dataset with multiple types of multiple ingredients (salt1,salt2, etc). Each n-th variation of each ingredient vs flavor may be represented by an n×k matrix that where an ingredient ...
1
vote
1answer
38 views

Is it ok to keep a very strong predictor and other weak predictors in the model? The model built is GBM

Age is coming out as a really strong predictor compared to other variables. This is a classification problem, the dependent variable is a (0/1)
2
votes
1answer
19 views

How can I compare the predictive power or association of two variables of different nature?

I am dealing with the following problem: We have 3 variables: A continuous variable (0 to 1), that is a scoring for people. A discrete variable, offered by a partner, in the range 1..10. That is also ...
0
votes
0answers
14 views

How to evaluate prediction error from population?

I have data of 30 subjects. For each subject, measurements have been performed every 5 minutes. Due to the different length of the duration of the procedures, each subject has different amount of ...
1
vote
0answers
45 views

Using Cross Validation for Goodness of Fit for Competing Inferential Models

I have a project where I need to 1) perform inference: understand the role of predictors on the response through models (with my data I need to choose a model where I can reasonably assess how changes ...
0
votes
1answer
16 views

Hypothesis not rejected, what does that mean for expected value? [duplicate]

So I have a linear regression analysis where the confidence interval includes 0, therefore the null hypothesis that the intercept is 0 cannot be rejected. Does this mean that the intercept value ...
0
votes
0answers
9 views

Model type / method for predicting cost of building remediation

I'm trying to estimate cost of remediation per apartment. So far I had very little success with glm (with various link function and / or log transformation) and xgboost. Building has N apartments and ...
0
votes
1answer
66 views

Use CNN to forecast time series value accuracy problem [closed]

I would like to use a CNN to predict a value based on some historical data. The concept is easy: I have a numerical value (label) the depends on some other numerical values (features). Each set of ...
1
vote
0answers
22 views

Prediction vs. model stability in inner and outer loop of repeated nested cross validation

Imagine I want to optimize some hyperparameters and get an estimation of the generalization error to compare different prediction models. I use a nested k-fold cross validation to avoid data leakage ...
2
votes
1answer
23 views

Entropy based measure of explained variation

I have an observable $Y$ which is a function of some set of variables $\mathcal{X}=\left\{ X_{n}\right\} _{n=1}^{N}$. Now $Y$ is a deterministic function of the $X_n$s, but conditioned on only a ...
0
votes
1answer
35 views

How exactly should repeated measures situation be treated in machine learning models?

I was reading this SE about repeated measures and I was a little bit confused. I have a data set consisting of claim information for million locations for up to 5 years and 99.9% of the observations ...
0
votes
2answers
36 views

How to interpet this equation? [closed]

Can I get some help on interpreting this equation? Is it saying which ever section separated by the commma is bigger would be the answer? How do you intepret Sigma pi= 1? Thanks!
1
vote
0answers
48 views

Diebold-Mariano test for predictive accuracy with R

The following example code is from the documentation of the command dm.test from R. ...
1
vote
0answers
14 views

Predicting future sample using Posterior Predictive Distribution

I have an observed datasample $X=[x_1,\ldots,x_t]$ from a time-series that is distributed according to a parametric model (Weibull) with parameters $\theta$. As I observe more data I update my ...
0
votes
0answers
15 views

Predicting if an incident will happen in an imbalanced data set

I want to know the best approach to tackle the following problem: We have row data for people involved in major incidents, and about 10% of people will have had a major incident at some point in the ...
0
votes
0answers
26 views

should we include or exclude a variable in a logistic regression based on the description below?

should we include or exclude a variable in a logit regr. model which will only obtain values if a certain event takes place otherwise will show N/A? this variable tells whether or not a product will ...
0
votes
0answers
24 views

Obtaining distribution parameters from GLM

My question is part statistical and part coding. Say I have fitted a gamma log-link GLM. doc example ...
1
vote
2answers
68 views

How to identify the most impactful features in a ML model, i.e. the predictor variables that can drive the biggest change in the target variable?

I have built a machine learning model using Random Forest in Sklearn (RandomForestRegressor). The model has up to 473 predictor variables and 1 target variable (all predictor and target variables are ...
1
vote
0answers
9 views

How to make predications based on past pair-wise comparisons

How can I predict the score that a non-present observer would have given, if they were present, based on past observations? I have five participants/observers (A B C D E). They are required to watch ...
0
votes
1answer
14 views

How to estimate model coefficients for option-like response: ie. response is related to variables in the form y = max( formula, 0)

Rewrite Hopefully someone can point me to a resource on how to estimate the parameters I'm trying to model. I've had trouble giving a title to my question and googling for resources. Suppose $y$ ...
1
vote
0answers
30 views

Is it necessary to test every assumptions for each predictor in a MLR?

If I have a dataset with several potential predictors and I want to perform a multiple regression model, do I have to test every assumption (normality of residuals, linear assosiation, ...
0
votes
0answers
10 views

how to interpret lifelines package output for survival analysis in python with normalized variables

I'm trying to interpret the result of a survival analysis that I made using lifelines package in python. I have normalized my variables to 0 to 1 range.I'm finding it difficult to comprehend. The ...
0
votes
0answers
22 views

Equalize multiple unevenly spaced time series for forecasting

I am building a time-series forecasting model to predict some patterns in climatological data. The dataset consists of many (2 mln) time series which look for example as: However the observations ...
1
vote
2answers
64 views

Compare predicted versus actual outcomes in a GLM

I read somewhere that you could compute a "residual value" for a GLM by taking the actual values of your response variable divided by the predicted value of that response variable. For example, ...
1
vote
1answer
53 views

Can I use linear regression analysis if the predictors and criteria change during time?

Can I use linear regression analysis to analyze the relationship between two variables if both of them are changing in time (each day)? Predictor = Number of unemployed people (in a country) ...
1
vote
0answers
15 views

Integrating logistic regression with CA to simulate land cover after prediction

I have used logistic regression in R to predict the probability of change in 3 types of land cover, the map of probability is generated for my case study. I do not know how to use the probability ...
1
vote
1answer
32 views

Why is the model performance better with more data, while it does not seem to be due to reduced model variance?

So according to most of the sources I saw, increasing training data size will only benefit high-variance ML models by exposing the model to less spurious patterns, which occur more frequently in ...
2
votes
0answers
16 views

What is an appropriate statistical technique to use with rates of decay/growth for this estimation problem?

I have the following problem: I have ground truth data on the population growth and the population decay of a certain breed of rabbits (let's say $R$) from $T_1$ to $T_n$. Now, I want to estimate ...
0
votes
0answers
34 views

Calculating Prediction Intervals for multistep univariate time series forecasting using Bootstrapping

I understand the way to compute the prediction interval at 5% and 95% for one step forward forecast based on Bill's answer to the question at Bootstrap prediction interval. The idea being that based ...
0
votes
0answers
6 views

One step prediction of time series using LSTM

I want to predict stock prices using LSTM. I have successfully trained my model and have it saved. Now that I've loaded it back in, how would I use model.predict() ...
0
votes
0answers
8 views

How can the prediction of a model be assessed?

I just played around with the VGG16 and ResNet56 model trained on the ImageNet dataset and realized, after running some tests, that the prediction confidence of both networks is really high even if ...
0
votes
2answers
21 views

Most likely sources of divergence between (adjusted)-R squared and out-of-sample predictive performance

I'm wondering which invalid assumptions are most likely to explain the wild discrepancies between a model's R-squared as a measure of predictive performance, and the actual out-of-sample predictive ...
1
vote
0answers
33 views

Using quantile regression results to select and weight variables for models

Linear regression is commonly used to identify predictor(s) (e.g., scores on cognitive ability or personality assessments) of job performance. Typically, predictors that exhibit a significant ...
1
vote
1answer
31 views
1
vote
1answer
22 views

Can we express the following unconditional probability as follows?

Some of you may be aware that I have been asking a nagging question for quite a while on this forum, in different shapes and forms. Although I may have been a nuisance, may I thank you as this has ...
0
votes
1answer
20 views

Isn't a simulation a great model for model-based reinforcement learning?

Most reinforcement learning agents are trained in simulated environments. And the goal is often to maximize performance in this same environment. Why is the simulation not used for planning in these ...

1
2
3 4 5
44