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
0answers
22 views

Transforming the independent variable for better prediction results with decision tree and random forest [duplicate]

I am trying to predict the amount a customer will pay. Removed zeros and standardized key columns. Data looks like this The distribution of a dependent variable is very skewed. I tried to apply ...
2
votes
0answers
22 views

Modeling Expensive Functions Using ML (Q1), Is $r^2$ Appropriate?

I may end up asking several follow-up questions in different posts, hence the broad view of my problem. Context and Background I have a function which is very expensive to calculate. Imagine ...
0
votes
0answers
10 views

Is curse of dimensionality more harmful for distance based models? [duplicate]

I want to understand what exactly is curse of dimensionality and how does it effect the model performance. Does the the concept apply to all the models?. Is it equally bad for distance based models ...
0
votes
1answer
17 views

Determining which categorical data is beneficial in predictive modelling

I am working on a model which will allow me to predict how long it will take for a "job" to be completed, based on historical data. Each job has a handful of categorical characteristics (all ...
0
votes
0answers
19 views

Model for predicting duration based on categorical data

I am working on a model which will allow me to predict how long it will take for a "job" to be completed, based on historical data. Each job has a handful of categorical characteristics (all ...
0
votes
0answers
23 views

What kind of prediction/ classification is this?

Suppose I have a data frame like the following (and I want to do machine learning): ...
1
vote
1answer
25 views

Prediction of group of values

I want to predict belonging to the range of values (e.g. age group) based on numerical labels (e.g. exact age). Is, in general, regression or classification more accurate to this problem?
1
vote
0answers
19 views

Create log likelihood function for three classes for SEIR modelling

I am working on SEIR model for measles data which has the following variables and sample observations. ...
2
votes
1answer
50 views

How do I use a distribution as an independent variable in a regression?

I'm trying to build a regression model that predicts Trump's vote share in a county in the 2016 election, given demographic data about that county. One of the demographic variables I would like to use ...
0
votes
0answers
16 views

Have I seemingly constructed a valid COVID-19 nearest-neighbor state forecast? [closed]

There is literature supporting nearest-neighbor model forecasting based on several favorable attributes. The first of which is apparent flexibility in choosing what “near” means in nearest neighbor ...
0
votes
1answer
20 views

How to validate a treatment scheme?

The following is purely hypothetical: 1. Treatments We have 4 treatments, let's call them A, B, C and D, for some medical condition, for which there is a 10% chance of recovery on average (given that ...
0
votes
0answers
18 views

Bayesian bootstrap predictive distribution

I have been reading this paper which introduces Bayesian bootstrap predictive distribution. In the text, author states that The Bayesian bootstrap predictive distribution is obtained by applying ...
0
votes
0answers
17 views

Gaussian Process Classification predictions

I am familiar with GPR and feel like I have a good handle on that, however the GPC still elludes me, specifically the prediction part. In Rassmussen Gaussian Processes for Machine Learning chapter 3 ...
0
votes
0answers
17 views

Pretrained models for sexual harassment detection [closed]

I'm looking for pretrained ready-to-use models for sexual harassment detection in chat messages.. Preferably (but not necessarily) free of charge, open source, and allowing for domain specific fine-...
3
votes
2answers
58 views

Dealing with groups of high dimensional data

I've got a dataset that follows patients who underwent different treatment options for aneurysms. They can have more than one aneurysm and each may be treated differently. So I have variables like: <...
0
votes
0answers
14 views

Possible to make predictions with overlapning data

I've the following dataset. The task is whether "UNEMP_Q1" is correlated with "UNEMP_AVG_Q47" and whether it is possible to use "UNEMP_Q1" (or/and UNEMP_Q2 and UNEMP_Q3 ...
2
votes
0answers
12 views

Application of xreg in forecasts

I'm trying to compare my forecast values with real values and assess the difference. I have 130 observations which are number of deaths (dependent variable). As well I have data for interaction ...
1
vote
1answer
34 views

Past price data has predictive power, it's not 50/50

They say that it's impossible (for ordinary men, with simple statistics and math) to predict future stock price based on the past stock price. But that's not exactly true. As you still can predict ...
0
votes
0answers
4 views

LSTM model convergence to a fixed state

I understand that this question has been asked previously https://stats.stackexchange.com/questions/253967/why-would-an-lstm-converge-to-a-fixed-state-when-generating-sequences?r=SearchResults&s=1|...
2
votes
0answers
36 views

How do I calculate confidence level or interval?

I have this data for exam grades in English Literature: 2019: the numbers achieving (A*, A, B, C, D, E) are (1, 5, 4, 7, 2, 0, 0) 2018: the numbers achieving (A*, A, B, C, D, E) are (2, 4, 3, 7, 0, 1, ...
0
votes
0answers
11 views

Determine Y using result from a Generalized additive model

I have a question regarding actually using the model created by the gam function from the mgcv package. Once I have tested my ...
3
votes
1answer
55 views

Logistic regression model predicts only one outcome, producing a high specificity but very low sensitivity. How do I improve the model?

I'm designing a logistic regression model to predict hospital mortality. Why? To identify 'adjusted' odds ratios for a variable of interest on mortality. Methods: - set up using a training dataset (75%...
0
votes
1answer
25 views

How to forecast with certain conditions

I have a doubt about forecasting analysis. Let's say I have a "time" variable and a "budget" variable. I would like to build two model: time series model where I forecast the ...
0
votes
0answers
30 views

How to develop a likelihood based prediction model to predict chance of rain in a particular hour of a year?

I have time-series weather data (from 2005-2018) of temperature and rainfall (every one-hour interval) from three different location weather stations. I want to predict/see what likely be a chance of ...
1
vote
0answers
17 views

Bibliography for Retail & marketing Analytics

I'm looking for some sources about statistical techniques than can be applied for marketing within retail industry (e.g. Coca-Cola, Pepsi, Heineken, and so on) I'm truly interested, by example in Key ...
8
votes
0answers
128 views

How would econometricians answer the objections and recommendations raised by Chen and Pearl (2013)?

In their article, Chen and Pearl (2013), critically examined 6 econometric textbooks, among these the textbooks written by Wooldridge (2009) {the introductory book}, and Stock & Watson (2011). ...
1
vote
1answer
13 views

How to compare importance of different predictors in different situations?

Suppose I have data at the individual student level on different components of standardized test scores (e.g. verbal, math, and writing components of SAT), college major chosen, and performance in the ...
2
votes
2answers
107 views

Predict daily count of delivered packages

I have a historical information for delivered packages (send time and receive time). Also I have an information for packages in transit (send time only). For example, all of the packages in transit ...
0
votes
0answers
13 views

How to perform time series prediction for small time series in rstudio using gam model?

I have a dataset which contains year(4 years - 2016, 2017, 2018, 2019), yield, ndvi, ndre,... savi indices, latitude and longitude of a field area of 8 hectacre. I have to predict the yield potential ...
0
votes
1answer
33 views

How to use a variable for training a model but not for making predictions?

I’m trying to predict the click through rate (CTR) of a product listing. As an input to train my model, I want to use the position of the product in the listing (if it’s the first product listed, or ...
1
vote
0answers
21 views

Cost sensitive learning and class balancing

I am facing a classification problem with classes that are really imbalanced (more or less 1% of positive cases). In addition, the "cost" of a False Negative (FN) is much higher than the ...
0
votes
0answers
11 views

Good references for 'predictive models'? [duplicate]

I am a recent graduate working as a Data Scientist and I often get lost in 'choosing' which model/models to use for a predictive task. Just recently I was trying to build a prediction on time-series ...
0
votes
0answers
13 views

Partial Rank Correlation Coefficient and R2?

Is there a R2-equivalent in using Partial Rank Correlation Coefficients (Pearson or Spearman) to get an overall predictive power of the multiple regression model? I'm wondering if there is one value ...
0
votes
0answers
24 views

Stacking and Ensembling methods in Data Science

Recently it seems that stacking and ensembling methods have become more popular, and using these methods can give better results than using a single algorithm. My question is: What are the reasons, ...
0
votes
0answers
12 views

How to fit simulated and measured data having systematic errors?

Consider a given time series of measured data and consider a simulation - consisting of a model, possibly fed with some parameters estimated a priori (not via the measured data) - which gives another ...
0
votes
1answer
34 views

is there a way to train the models in batches? i m not talking about deep learning

I have 800,000 records. The training with all the records at once is not possible. Is there a way i can train a model on 50k records then continue training on next 50k records and so on... basically i ...
3
votes
2answers
69 views

Why can't we mix exploratory model building and statistical tests for validating a known model?

This short article attempts to expand on a lesson that was drilled in to me during my stats courses. That is, the point that some sorts of tests should never be mixed with ANOVA. It even contains this ...
0
votes
0answers
27 views

Can one make meaningful predictions from “censored data”?

In an ML project, I'm using patients' clinical and demographic features to predict their treatment outcomes. Among over 300K records, only 6,022 are known to have experienced severe conditions (ICU, ...
0
votes
0answers
23 views

Feature selection and cross-validation without external validation

I have a microarray panel of > 40 marker antigens for more than 500 people, 400 of them being symptomatic controls in the gold standard. Each patient The blood was drawn each time individually at ...
3
votes
0answers
26 views

Bootstrap Resampling vs Repeated K-Fold Cross-Validation for Confidence Intervals

Why is bootstrap resampling with replacement used to construct confidence intervals over repeated K-fold cross-validation? Isn't it valid to use 10-fold CV repeated 10 times, where we garner 100 data ...
1
vote
1answer
39 views

Multi-class classification with prior knowledge of class similarity?

Backrounds I would like to build a model that predicts a month label $\mathbf{y}$ from a given set of features $\mathbf{X}$. Data structure is as follows. $\mathbf{X} : N_{samples} \times N_{features}...
1
vote
0answers
16 views

handling multiple time series through common model?

I have 1.5 lac/ 150 K timeseries . These are divided by geo locations. I have total 32 geo locations.Customer is expecting to have minimum number of model for all the 1.5 lac forecasting. How should i ...
0
votes
0answers
4 views

Extrinsic Evalutation Peformance of Google Autofill

Does anyone know approximately how successful Google's autofill algorithm is for search completion? Does is successfully return the next word the user was searching for in it's top 5 suggested results ...
3
votes
0answers
47 views

multicollinearity and predictive power

In wikipedia it says ...
1
vote
0answers
21 views

Interpreting GAM plots: gam.check()

I am new to building GAMs, and would appreciate any help interpeting the graphs produced by gam.check I have written a gam model to predict probability of "success" in growing a species of ...
0
votes
1answer
19 views

refitting a model to the data during a backtest

I built a back-tester for sports betting, I'm not sure whether to add a re-fitting feature. As in, while the back-tester is evaluating the model out of sample, should it be able to re-fit to data that ...
0
votes
0answers
11 views

Month over month prediction for credit defaulters

I'm working on a month over month prediction using credit historical data. I've created more than 600 variables for this prediction including customer's delinquency in last 3/6/12/24 months etc. I've ...
2
votes
2answers
109 views

Linear Regression for Noisy Data

I have noisy dataset collected from a source and I am planning to fit a regression to this dataset. The dataset has Y and X1 variables (both continuous between (-1, 1)) and I plotted a scatter plot to ...
0
votes
0answers
13 views

Query on Precision-Recall curve

I have an imbalanced dataset - Kaggle's Porto Seguro insurance dataset. I have applied Random Forest and XGBoost classifiers on the imbalanced dataset, under-sampled dataset and over-sampled dataset. ...
0
votes
0answers
17 views

Best practices for predicting trial to paid conversions for in-progress trials

I have a data frame that has tens of thousands of rows that look like this: ...

1
2 3 4 5
44