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
0answers
34 views

Why ever use any of the tests of hypotheses when we can just regress everything?

So far in every data set I've worked with, regressing with only 1 predictor variable has yielded p-values extremely close to the one I get when running it with a test of hypothesis such as pearson's, ...
3
votes
2answers
128 views

How to predict the degree to which an extraneous variable will attenuate a correlation?

Assume there is a predictor x (a video-recorded job simulation) that correlates r=.3 (Pearson r) with a criterion y (later job performance). Assume a new grading process is used and it is noticed ...
4
votes
2answers
143 views

Cox regression vs logistic/simple regression

I understand cox regression is used to calculate a risk ratio while logistic regression calculates an odds ratio. Does that mean that in retrospective cohorts I would use cox regression while in ...
1
vote
0answers
65 views

How to make sense of logit summary output?

I ran a logit model using statsmodel api available in Python. I have few questions on how to make sense of these. What's the difference between summary and summary2 output? Why is the AIC and BIC ...
1
vote
1answer
30 views

Mixed models question

Let's say that i have data with 5000 participants(rows) and their scores on some sports, their age, weather on each event, location etc. Is it appropriate to use linear mixed models(lmer in R) if i ...
2
votes
1answer
229 views

Higher RMSE lower MAPE

I have a time series model that forecast next K days. For example when I forecast next 50 days my MAPE is 20.3% and RMSE is 2943 and when I forecast next 200 days is the MAPE is 10.25 % but RMSE is ...
0
votes
0answers
24 views

What does Maximum Likelihood Estimation mean in Machine Learning? [duplicate]

I am wondering what Maximum Likelihood Estimation means to Machine Learning in terms of training a predictive model. I understand Machine Learning uses Maximum Likelihood Estimations for model ...
0
votes
0answers
53 views

Regression: Can I use as outcome variable a score that is the total score of all the predictor variables?

I have a data set of decathlon sports which contain the performance of each one of the ten individual sports and then the overall score which is the summation of each one of the individual sports. ...
1
vote
0answers
130 views

How to measure Sales lift for a campaign without experimental design?

I want to be able to calculate Sales Lift for a campaign conducted but prior to campaign no control group was established. So, I cannot measure the impact on Sales because of campaign to treatment ...
0
votes
0answers
9 views

Repeated cross-sectional data: predict later scores based on 7 earlier measurements

I have data from 22 organisations and repeated measurements with individual responses within each organisation. There are up to 7 measurement occasions, each a year apart. But I do not have ...
0
votes
0answers
19 views

Finding out whether events are correlated

I manage software development. We have a lot microservices and have built a lot of features. I would like to see if any of the microservices are highly coupled - which I'm defining here as meaning ...
0
votes
0answers
45 views
3
votes
1answer
272 views

Get equation from glm coefficients: calculate y manually?

I am trying to understand the math behind the glm(). Specifically, how to apply equation based on model predictors to calculate my ...
0
votes
0answers
21 views

Using Self-Starting Functions For Exponential Decay Rates

I am trying to decide how to get my starting parameters optimized for a simple exponential decay model. My regression formula is: y ~ Be^(at) Where y is the remaining value at time,t, remaining ...
1
vote
1answer
58 views

To determine variables to figure out the bad customers in credit risk modeling [closed]

I am developing a probability to default model on a data from landing firm. After running the GLM() model i have got the below message: ...
0
votes
1answer
28 views

RNN model for predicting room temperatures

I am currently doing a project in Machine Learning where I am trying to predict the temperature of a room in future. I have a 1-year dataset of a house with 12 rooms. Data is collected at 10 min ...
0
votes
0answers
14 views

Nested Cross Validation: Choosing a Classification probability threshold

With nested CV: Inner loop for model selection, outer loop for performance evaluation. At what level can we optimize a threshold probability (vs. 0.5) to maximize sensitivity or specificity of the ...
2
votes
0answers
56 views

finding different value while doing regression

My question may sound unclear but I'm going to explain it in detail. I have a data set includes two-column with 50 samples. let's call the first column energy and the second column cracking. Let's ...
1
vote
0answers
12 views

Identify Influential predictors

I have a dataset with binary class as outcome. I was exploring the data through plotting the variables for both the classes. For example, something like below ...
0
votes
0answers
25 views

What does prediction mean in Bayesian Network and how can I make predictions in Bayesian Network?

I am currently working on a Bayesian Network(1 parent node, 2 children nodes, for example) and want to do predictions on my Bayesian Network. Conditional probabilities are all set and I wonder the way ...
0
votes
0answers
39 views

Output of xgboost() while optimization is not very intuitive

I am running xgboost() on a data set with a data set with below columns. ...
3
votes
1answer
96 views

Filling missing data points with lmer prediction model

I'm trying to interpolate the missing data point using lmer model prediction. Subsetting to a table without any na to the missing column of interest: ...
1
vote
1answer
26 views

How to find significant predictors that can differentiate case and control without ML approach?

I have a dataset with more than 70 columns and I have an binary output column. What I did currently was to explore the dataset by plotting the bar and line graphs for the input variables vs output ...
1
vote
1answer
30 views

How to manage a variable collected at various level in a machine learning model based on nested hierarchical data?

I'm trying to use machine learning to model the risk of healthcare-associated infections (HAI) for patients in a number of hospitals. I have variables both at the patient, ward and hospital level. ...
0
votes
0answers
7 views

What is best practice when standardizing a truncated numeric variable with lots of zeroes?

What is best practice when standardizing truncated numeric variables with lots of zeroes (like 80% of the obs.)? To provide an example, I have a variable counting number of days per year several ...
0
votes
0answers
34 views

Imbalanced data classification with GLM giving very poor results [duplicate]

I have a loan defaulters dataset and it is highly imbalanced as shown below: 0 1 33108 673 I have tried SMOTE to balance the dataset, as shown below: ...
0
votes
1answer
43 views

Are the forecasting methods like mean, naive, drift, weighted average applicable to non stationary time series?

Like AR, MA models essentially need the series to be stationary, do the other forecast methods mentioned above also follow stationary?
2
votes
3answers
88 views

What methods of forecasting should I be looking at to forecast sales?

I am wanting to forecast sales of different products within a business. I have a good background in mathematics (but mainly focused on analysis, group theory, algebra etc. as opposed to statistics). ...
0
votes
0answers
12 views

Defining an applicable score index for patient survival prediction

Say I want to devise a score index for predicting patient overall survival (OS) based on a handful of co-variates, what would be the best way to do it? Briefly, I have performed a Cox regression ...
0
votes
1answer
36 views

Is there a way to ensure LASSO regularisation retains certain features in R? [closed]

I am creating a predictive model with a large number of features in R, but would like to prevent basic demographic features from being selected out of the model via LASSO regularisation. Is there a ...
2
votes
0answers
230 views

Using training set / validation set for prediction after model assessment with test set?

When there is enough data, ML splits the data into 3 sets: (1) training, (2) validation, and (3) test set. So in the training stage, an algorithm is trained in (1), "validate" the trained model's ...
0
votes
0answers
36 views

Predicting the loan default probability for next t intervals by predictSurvProb()

I have a loan dataset that contains details of customers for different loan types. Every category of loan have different tenure(starting from 6Months to 60Months) We are defining "Good" or "Bad" to a ...
0
votes
0answers
13 views

Avoiding bias when labelling to construct a test/train data set

I have a multiclass classification problem, where I have a massive database of text documents and the objective is to predict what class a document belongs to. There are many classes (>10) but the ...
3
votes
1answer
64 views

Is a regression using a categorical variable a multiple regression by definition?

When running a regression with a categorical variable as the independent variable, the regression essentially picks one of the levels to leave out and runs all the other levels together. There is no ...
1
vote
1answer
89 views

Defaulters prediction on next cycle

I have data of loan installment repayments by customers, it contains all regular details like loan amount, last installment paid amount, next installments, credit score, age, region etc. Along with ...
0
votes
1answer
29 views

Predict satisfaction score given a shift of Service Level Agreement

I would like to make predictions for Overall Satisfaction based on a shift of Service Level Agreement (SLA). I have number of days taken to complete a single ...
0
votes
2answers
113 views

Predictive performance of joint models versus standard survival models

I am trying to show that predictions based on repeated measures of markers (using joint modelling of repeated markers and time to event models: JMbayes package) are better than those based on only one ...
4
votes
1answer
414 views

Cross Validation and Multiple Imputation for Missing Data

Using 10 fold CV for performance estimation of a logistic regression model, what is the appropriate way to incorporate multiple imputation for missingness across the predictors and outcome in which ...
1
vote
0answers
31 views

Calculate CLV / TLV for contractual Business

I am desperately trying to apply a CLV/TLV (Customer Lifetime Value) algorithm to my dataset in R. Unfortunately, the more I read about it, the less confident I get if all makes sense. =) Do you have ...
0
votes
0answers
21 views

How consistent should a neural network's results be after repeated training on the same dataset?

For example, I can train my network on the training data and achieve a prediction accuracy of 54%, but then do it again with the same architecture and dataset and get 55%. Should I be tweaking my ...
0
votes
0answers
13 views

What kind of models can handle information coming in or changing over time?

Apologies if this question is too general - I can try to be more specific if needed. I'm trying to predict the outcome of certain incidents. The target variable is $Y$ (in some cases its categorical, ...
0
votes
0answers
6 views

Predictive model for product-wise monthly sales growth, large unbalanced panel data

I have over 30,000 individual products and sales data by month dating back to 2013. Many have been added in between now and then and i want to aggregate all the sales to create a predictive model for ...
1
vote
0answers
14 views

Properties of the mean of a predictor function

I am learning about the bias-variance tradeoff, and my question is about the properties of the function $E_{D_n}(\hat f(x)) $ which appears inside both the $variance$ and $bias^2$ terms. I am looking ...
1
vote
0answers
36 views

Posterior Predictive Checks in R using rjags

I am working on a problem that involves the following information. I have managed to answer the first part of the question using the following r-code that allows me to compute the posterior ...
3
votes
1answer
118 views

Reconstructing a logistic regression model from literature using published coefficients

I have a logistic regression model with the form: logit(p) = alpha + X*beta Where alpha is the intercept, X is the covariate matrix, and beta the corresponding coeffiecient. I want to be able to ...
1
vote
0answers
20 views

Unrealistically high AUC-ROC score comparing to control feature and other performance measures

I am making a binary classification using regularized logistic regression, with extreme unbalanced data. The target label is Tar and non-target label is ...
0
votes
0answers
18 views

Perform cross-validation for variable selection where “final” model must include certain variables in R

I have a data set with 20 predictors and a dependent variable. My goal is to perform k-fold cross-validation (CV) to select the set of most predictive/relevant predictors. My question is: Supposed ...
1
vote
2answers
289 views

What is the most appropriate way to validate prediction models with clustered data?

I am attempting to develop and validate a multivariable classification model using data from 10 clinical trials. I would like guidance on the most appropriate way to validate (internally and ...
1
vote
1answer
30 views

Posterior Predictive CARBayesST

I'm trying to use the CARBayesST package and I need to do Spatio-temporal predictions. In the vignette of the package on page 27 says " If there had been saying m missing values, then the Y component ...
0
votes
1answer
35 views

How to deal with categorical independent variables with numerous levels [duplicate]

How to deal with regression when most of the independent variables are categorical having numerous (more than 10) levels and the dependent variable is continuous? For this would it make sense to ...

1
3 4
5
6 7
44