A message from our CEO about the future of Stack Overflow and Stack Exchange. Read now.

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
11 views

How to predict future data using a sample

Is there a statistical tool to create a confidence interval of sorts that instead of trying to pin down the population mean it tries to pin down the mean of a future sample. In the same way that a ...
7
votes
1answer
134 views

Recurring problem with retrospective data collection study designs I'm seeing

I've noticed a lot of medical research that I am involved in goes as follows: Collect data on 300-1000 patients, including all sorts of baseline characteristics such as BMI, age, gender and then ...
1
vote
0answers
12 views

Commonly-used Predictor Function?

I was recently taking a look at some political projections (similar to the ones published by 538), and I am wondering about a particular function used to calculate the likelihood of a given party ...
0
votes
1answer
26 views

Aggregating binary data into a “score” reduces CI…Does it increase the validity?

I have 5 variables that I've used to calculate a score with. Each variable contributes 0-2 points to the score depending on their value. I'm interested in whether these predictors better can predict ...
0
votes
0answers
34 views

Is it possible to inverse a Black Box Model with a Neural Network?

I have multiple sets of data from previous simulations, input parameters {P1, P2} and corresponding output curves {y1, y2, ..., yn}. The simulation's model is expensive to compute, so I would like to ...
2
votes
0answers
16 views

Impact of substituting default values for predictors in Regression Model

Suppose I have a regression model (it could be any model, but I am mainly interested in multiple linear regression) that is designed to predict Y as a function of a set of X predictors. In my field (...
0
votes
0answers
15 views

Regressing predictions on corresponding observed values, does that make any sense? If so, would this be a reasonable proposition?

I have a list with around 100 different dependent variables, for each of them I have around 35 observed values with 12 explanatory variables (it’s the same 12 explanatory variables for all dependent ...
1
vote
2answers
18 views

Appropriate statistical analysis for predicting early death or early progression

I am working on an analysis seeking to predict early death which is defined as death within 3 months of the start of treatment. What is the best way to analyze this since there is censoring involved ...
13
votes
4answers
894 views

Dichotomizing continuous variables at their optimal cut-off for clinical interpretation

In medical context, when presenting results from a binary outcome with a continuous predictor, the OR (odds ratio) can be difficult to interpret. Example: A doctor does a study in which he wants to ...
0
votes
0answers
21 views

What does it mean if a regression model is not significant, but individual variables are? [duplicate]

With a logistic regression model of 3 independent variables, one of the variables is significant while the other two are not. The significant variable is my variable of interest while the other two ...
0
votes
0answers
13 views

How to compute Net Reclassification Improvement or Alternatives?

I am working on a binary classification problem with ~5k records and class proportion of 33:67. I have 60 features/variables in my dataset and finally I have come to about 10 variables based on ...
2
votes
0answers
30 views

200 unique respondents, over 3000 observations?

If doing a survey study with 200 respondent resulting in over 3000 observations (respondents answer several questions) how do you approach this with a regression? We've split the data into segments ...
1
vote
1answer
32 views

How to compare and evaluate models for a new feature?

I am working on a binary classification where I have 4712 records with Label 1 being 1554 records and Label 0 being 3558 records. When I tried multiple models based on 6,7 and 8 features, I see the ...
1
vote
1answer
27 views

Correlation being muddied by outliers

I have a study in which I find a decent correlation: on a quadratic prediction plot between a binary outcome and a continuous x. However, there are a few observations that have numbers that are not ...
2
votes
2answers
31 views

What can spearman's rank do that regression can't?

It seems to me, that with today's computing power, tools such as spearman's rank/correlation are completely useless. They uncover the exact same information as a regression, except they can't make ...
0
votes
1answer
26 views

Low N, unsure on what model to build [duplicate]

I've got a study with 55 patients having undergone surgery. 80% were happy with the surgery while 20% weren't. I'm looking at predictors that may be able to predict surgery satisfaction. The problem ...
0
votes
1answer
31 views

What counts as a comparison?

Say we have a test in which we are comparing many variables to a dependent variable, using different methods. Say ttest, chi2 and then finally a multiple regression on parts of the data. Would we ...
1
vote
1answer
33 views

Which model to choose for time of day as a dependent variable having a skewed periodic/rhythmic relationship with some response?

I have a data-set with time of day (0 - 24 hours) as a dependent variable together with some continuous response variable which demonstrate what looks like a skewed sine relationship. For ...
2
votes
0answers
35 views

Getting started with regularization (Lasso)

I've got a small data set of 55 observations with a binary outcome variable of which only 11 are 1's and the rest are 0's. I was wondering if Lasso was a useful tool to predict my outcome here and if ...
1
vote
0answers
18 views

Would a multiple regression controlling for one variable increase the validity of my study?

I have a study in which the problem is a very low power (low amount of people in the outcome variable). I found during a previous question I asked that we should probably not use multiple regression ...
1
vote
1answer
37 views

Having a problem reading my multiple logistic regression model

I have a logistic regression model that is seemingly significant when regressing individual variables in a univariate regression, but the entire thing falls apart when input into a multiple model. I'...
1
vote
1answer
38 views

Using multiple regression to control for variables

If you have a variable you want to investigate but you realize you should probably control for a number of variables, is it okay to do this for every variable you find interesting? Say you have 5 ...
0
votes
0answers
13 views

Getting extraordinarily large coefficients on otherwise seemingly normal numbers in logistic regression?

I've got a logistic regression for a binary outcome and a continuous predictor that stretches from 0.28 to 0.55. The relationship is U curved in nature so for predictor "p" and dependent "d" I regress ...
0
votes
0answers
26 views

Modelling a quadratic relationship

When reporting a quadratic relationship (U curve) in a logistic regression, what exactly are we doing? I understand the formula for the quadratic function is $a_1x_1^2 + a_1x_1 + B_0$ but that's ...
1
vote
2answers
86 views

Bayesian Linear Regression Predictive Distribution

This is a homework assignment. I need a hint in the right direction and a couple explanations as things are not very clear to me. What is the predictive distribution? How do I draw from it? The ...
0
votes
0answers
13 views

Tagging to similar data points

I have a data set, where customers who are non-delinquent with us, have defaulted with others. So from total data-set 30% are those who defaulted with others but 70% are those who are good with us ...
1
vote
2answers
45 views

Moving away from stepwise type of model building

I do a lot of studies in which we have a disease/outcome and then we collect a lot of information on the patients such as age, gender, BMI, comorbidities, lifestyle factors etc. and then we run a ...
1
vote
1answer
43 views

When to drop correlated features?

I have a dataset focused on binary classification with 60 features and 5k records. Am trying to 1) find the risk factors using statsmodel logistic regression (I do this because it's important to ...
1
vote
2answers
123 views

Why does the constant term have a P value in statistical programs?

I notice when running regressions in programs such as Stata that the constant term has a p value. Why and what does this p value represent?
3
votes
2answers
255 views

nonsensical intercepts for regression models

Let’s say that I have performances in 9 sports as predictor variables and the total points of those sports as the DV. So I am making three regression models(non-nested) with three predictors each (...
2
votes
1answer
47 views

Understanding the 10:1 events per variable rule

I read online that 1:10 rule is based on the frequency of lower occurring class. I have a dataset with 4712 records. There are 1558 records labeled yes, and 3554 records labeled no. In my case, the ...
2
votes
1answer
64 views

Comparing different model with the same output

I work on developing 3 different models for predicting land surface temperatures in an area (validation is done with images taken by a thermal drone). The models are: A thermodynamic mechanistic ...
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, ...
2
votes
2answers
115 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
113 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 ...
0
votes
0answers
14 views

Will removing zero variance or correrelated predictors prior to splitting into Train and Test sets introduce bias or data leakage?

My question is about bias and data leakage when building a predictive model. I understand that any scaling or standardization should be done after a Train/Test split of the data is performed. Is the ...
1
vote
0answers
18 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
28 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
21 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
20 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
50 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
14 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
6 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
18 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
35 views
3
votes
1answer
63 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
6 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
49 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
19 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
8 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 ...