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
0 votes
0 answers
8 views

When should I balance my data using AUROC and AUPRC?

I want to report the AUROC and the AUPRC of a prediction model using an unbalanced dataset. Is it correct that I have to balance my data to calculate the AUROC but leave the data unbalanced to ...
user avatar
  • 21
1 vote
1 answer
18 views

Given a specific value for a variable, how do you find the predicted value of a fixed effects multivariate regression?

I have specified the following model in R: model <- lm(yield ~ N + N^2 + P + K + S + factor(year) + factor(variety), data=data) As shown above, I ...
user avatar
  • 11
0 votes
0 answers
14 views

What is the proper way to compute predicted probabilities in a fixed-effects logistic regression?

Predicted probabilities are quite helpful to interpret the output of logistic regressions. One can compute a predicted probability as follows: $$ P = \frac{exp(\beta + \beta_0)}{1 + exp(\beta + \...
user avatar
  • 187
0 votes
0 answers
12 views

Post analysis using raw data or SHAP values in Machine learning

Let's say I have SHAP value returned in dataframe for input variables like below ...
user avatar
  • 1,758
1 vote
0 answers
41 views

Significance test for comparing different 10-fold cross-validated Machine Learning Regressions

Is there a recommended significance test for comparing different 10-fold cross validated regressions? For instance, I want to compare the performance of LASSO against Random Forest for my dataset. ...
user avatar
1 vote
0 answers
10 views

Predicting Value in multi-dimension data

I need help with my data, I mix 2 different powder materials to get one powder with a specific quality number. So, this resulted in quality numbers depending on 4 features, the quality of materials 1 ...
user avatar
1 vote
1 answer
68 views

When there are many more controls than cases, can I take only part of the controls?

I've tried to build a predictive logistic regression model. However, there are only 500 observations with disease (+) and over 60,000 observations with disease (-). Can I take a random sample (e.g., ...
user avatar
0 votes
0 answers
7 views

Variance explained at each level of a categorical variable

I have a deep learning regression that predicts the value of a continuous variable Y. There is a categorical variable Z that has ...
user avatar
  • 4,562
0 votes
0 answers
18 views

Missing "None" class in outcome variable. Least bad way of handling missingness?

I'm dealing with a preexisting dataset with an outcome variable of suicide which entails the following classes, of which multiple can be selected, but they roughly escalate in severity. Check if ...
user avatar
0 votes
0 answers
11 views

Why variable representation plays a role in prediction?

I am working on binary classification using a random forest, where the data have 977 records and 6 columns. The class ratio is 77:23. I have two derived input variables. One variable is called ...
user avatar
  • 1,758
0 votes
0 answers
41 views

Estimating Top n Prices points on a given day for a particular product which would maximize revenue

Problem Statement :- On any given date of the year, get top n prices which would maximize the revenue for that day for a particular platform. Dataset :- Date Price ($p_{i}$) Platform $X_{1,i}$, $X_{...
user avatar
  • 133
0 votes
0 answers
28 views

Using machine learning model trained on standardized data for real world low volume data

I have developed a machine learning model which has been trained on a preprocessed data by scaling and centering using h2o package of R. I am able to use this model ...
user avatar
  • 197
1 vote
1 answer
26 views

How to predict multiple future values in a linear model in R?

So i currently have a data set consisting of the Year, Credit Hours, and Number of students. I have been trying to predict future credit hours by the number of students. ...
user avatar
  • 11
1 vote
1 answer
12 views

How to label target dataset based on reduced dimensions of a source dataset?

I have a high-dimensional data matrix with K observations and N variables. To predict the label for each observation, I use some dimensionality reduction method (let's say PCA). Now I have K ...
user avatar
0 votes
2 answers
18 views

Intuition for confidence intervals vs prediction intervals for linear regression

I am having a bit of trouble understanding the difference between a confidence and prediction interval in the context of linear regression, and in what scenario we would use either of them. I've posed ...
user avatar
0 votes
0 answers
24 views

Difference between "predict" and "predictSurvProb" R functions

I am trying to predict survival probabilities after estimating a Cox model with coxph in R . I am aware of 2 functions: predict ...
user avatar
  • 583
0 votes
0 answers
22 views

Policy evaluation using machine learning techniques when interested in impact of certain covariates

I want to evaluate how policy changes affect a certain outcome. For this, I've built a model to predict the outcome in which, due to a high amount of covariates, I have used regularization. In a next ...
user avatar
  • 1
2 votes
3 answers
615 views

Negative prediction values from linear regression in R

So I made a linear regression in R Studio to predict the price of a car based on the year of fabrication. The data set is called "audi" and my linear regression looks like this: ...
user avatar
0 votes
0 answers
13 views

Measure for similarity of two distribution ECDF with sensitivity to the tails

Currently, I am looking for a measure to quantify the overall dissimilarity or similarity of two sample distributions (possibly of different size). I would like to compare observed data with model ...
user avatar
  • 11
1 vote
1 answer
42 views

Forecasting based on few samples

I have to forecast number of enrollments for an international univeristy , challenge is there are only few years of data.So, my data looks somewhat like this: ...
user avatar
0 votes
0 answers
30 views

With Stationarity How can ARMA Modelling have any Validity?

I have recently been thrown into the deep end with time-series econometrics. The first thing I have learned is that in order to avoid the spurious correlation trap, I need to ensure that all the ...
user avatar
0 votes
1 answer
64 views

Choice Between Alternatives in Machine Learning

I need some advice on the simplest/best way to structure an ML model for a (slightly) non-standard situation. Setup: I have many teams in a company that have leaders. Each team has two options for a ...
user avatar
0 votes
1 answer
50 views

Forcing covariates to always be part of a Lasso model

I want to use a Lasso to predict outcomes for different policy scenarios. At the optimal degree of regularization obtained by cross-validation, one important variable in whose impact I'm interested in ...
user avatar
  • 1
0 votes
0 answers
13 views

Deriving Bayesian Credible Intervals for AUC using R brms

I am trying to estimate the posterior distribution for the AUC of a predictive biomarker using R brms. However, whenever I calculate the AUC using the posterior distribution of the model parameters, ...
user avatar
  • 43
0 votes
0 answers
8 views

How to fix error in Decision tree prediction in R?

I am unable to fix the code for decision tree prediction. So, my dataset includes 20 variables. head(newdf) Then, I partitioned my data and created a decision ...
user avatar
  • 1
0 votes
0 answers
40 views

Developing a predictive model using only cross-sectional data

Let us assume I am trying to develop a predictive model that will give an indication of the progression of the percentage crack area on bridge decks. Engineering knowledge indicates that the crack ...
user avatar
  • 191
0 votes
0 answers
17 views

ggpredict produces flat fit line for glmmTMB model

I am conducting a step selection analysis of animal movement data using a Poisson model in glmmTMB. The model runs as expected, however, when I use ggeffects to plot model predictions I get a flat ...
user avatar
1 vote
1 answer
20 views

Prediction based on correlation

I know -- correlation does not mean causation. However, I need a "best guess" estimate for the following scenario: Variable A = Starting Point: 4,5 (SD=0,5) Variable B = Starting Point: 300 (...
user avatar
0 votes
0 answers
17 views

Under what conditions will prices of shares in a binary prediction market accurately represent probabilities?

I often see that prediction market sites say that the prices of the shares on outcomes can be interpreted as the likelihood of the outcome occurring. But under what conditions is this true?
user avatar
  • 1,048
3 votes
1 answer
26 views

Features are Relevant for Regression but not necessarily for Classification - what to make of this?

I have used the R Boruta package to check for feature relevance in predicting log returns of financial time series, the targets being the log returns themselves (for regression) and the sign of log ...
user avatar
0 votes
0 answers
22 views

From OLS regression to Logistic Model

I'm currently working on my master's thesis in finance. Without going into to much detail, my goal is to regress certain predictors on first-day returns (SPAC IPO performance). However, after ...
user avatar
  • 101
0 votes
0 answers
18 views

Predicting repeated binary response from contacts over time

From their initial entry date, people must be contacted each month for three months. It often takes several unsuccessful attempts before you establish contact. Once you reach someone, contact attempts ...
user avatar
  • 2,012
2 votes
1 answer
166 views

Logistic Regression on multiple classes (Shouldn't it be only on binary?)

I'm a bit confused with the usage of logistic regression for multi-class classification. My understanding is that a logistic regression is dichotomous (two possible classes), so in the example of the ...
user avatar
2 votes
1 answer
32 views

Can I improve linear model coefficient estimates using group information without working it into model?

I am fitting a linear model in order to predict future observations. The training data consists of about 1000 observations. Each observation comes from one of 10 individuals, which means I have about ...
user avatar
1 vote
0 answers
17 views

Metric that quantifies 0-centered error throughout the entire range of the data?

Say I am selecting between models, and I especially value mean-0 error throughout the entire range of the data. I am looking for metrics that specifically capture this property. For example, take the ...
user avatar
0 votes
0 answers
22 views

Marginal/Adjusted Predictions Python

I am an R user trying to learn Python. Is there a way to find marginal predictions of an effect in Python? What I mean is finding the prediction of a particular variable while setting all other ...
user avatar
  • 185
0 votes
0 answers
15 views

Predictive Distribution of Time series with Uncertain Future Values

In machine Learning, and especially in Turning Point Detection Problem, it is important to have the best estimate for the probability distribution function (PDF) of the future samples. Lets say that ...
user avatar
0 votes
0 answers
25 views

Should I use the predictions from my model or use the predict function on my model?

I have a small dataset (with about 5 covariates and 30 rows) and I am trying to make some predictions using R. I was advised to use Leave-One-Out Cross Validation (LOOCV) with a random forest due to ...
user avatar
  • 185
1 vote
1 answer
21 views

Logistic regression predictions dont work

I have this problem with logit, that when I want to create confusion matrix, it simply displays the real values in the first row and in the second row, there are never any numbers. I created a lot of ...
user avatar
  • 11
1 vote
1 answer
28 views

Ridge classification: Interpreting prediction

I'm particularly concerned about the following problem when using ridge classification for predicting binary outcome When I'm encoding the binary outcome as 1 and 0; my model accuracy is 0.6456 When ...
user avatar
  • 31
0 votes
0 answers
21 views

Logistic regression with counts over sliding windows of time

I am working on a logistic regression model that attempts to predict failure events in the following year over a lot of devices (say 1000) using the previous number of minor incidents that happen in ...
user avatar
  • 393
2 votes
1 answer
32 views

What reasons beyond interpretability are there to use additive models over a complex, multivariate smoother?

Let's adopt for the second the notation from the R package mgcv. ...
user avatar
  • 1,023
0 votes
0 answers
25 views

Ranking probabilities of logistic regression models M1 and M2 taking confidence intervals in account

I have two models M1 and M2 and each models the probability of having cancer with logistic regression. M1 is based on independent variables IV1 measured on a given sample of individuals and M2 is ...
user avatar
  • 393
2 votes
1 answer
121 views

Appropriate way to get cross validated performance metrics

For cross-validation of a logistic regression classifier, it seems to me that there are a number of different approaches to calculating each performance metric: The performance metric is calculated ...
user avatar
0 votes
0 answers
41 views

R Find the intersection of two lines for each group

Given the following data.frame: ...
user avatar
0 votes
0 answers
17 views

Estimating total number of consumers based on observed transactions

I read through most other questions of this type and didn't find a relevant situation. I want to predict the total number of consumers and repeat consumers that shopped in the past month, however, ...
user avatar
0 votes
0 answers
49 views

How does posterior predictive mean depend on parameters of the likelihood and prior distribution?

I have come across a problem in my research which deals with the mean of the posterior predictive distribution, i.e. $$p(x'|x)=\int d\theta p(x'|\theta)p(\theta|x)$$ where $x$ is an observed sample ...
user avatar
1 vote
1 answer
22 views

Accuracy failure example

I am trying to understand the functioning of accuracy and I need a practical example. This is what I understood: it gets the average correctness of the predictions and in some cases its result can be ...
user avatar
0 votes
0 answers
19 views

Regression analysis with max value

I am trying to find the best curve that will describe my data. my data are stored in numpy arrays of t and dur they are both in ...
user avatar
0 votes
0 answers
31 views

What is a model suitable for prediction of binary values based on time series?

There is a time series of stock quotes, length $n=500$. The data looks like: ...
user avatar
  • 794

1
2 3 4 5
90