Stack Exchange Network

Stack Exchange network consists of 174 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

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 ...

2
votes
1answer
27 views

Predictive distribution: What can we say about the prediction?

I am wondering what could we obtain more from the predictive distribution. Give a set of data, say $\mathcal{D}=\{(x_i,y_i)\}$, we want to predict the value s$Y_{new}$ at new locations, say $X_{new}$. ...
1
vote
2answers
55 views

Should I learn R for a project? [on hold]

I am in charge of an academic research project, building predictive models based on clinical data from multiple longitudinal cohorts of diseased patients. The total number of patients and controls (...
0
votes
0answers
8 views

How to predict the responses of survey respondents based on demographic training data using R

I have survey data collected from an online representative sample of some 2000 UK citizens. I ask each respondent whether or not they would support a certain tax policy, and each can answer either "...
2
votes
2answers
33 views

How does a fitted linear mixed effects model predict longitudinal output for a new subject?

I fitted a linear mixed effects model using nlme package for aids dataset. Here, CD4 is the CD4 cell count, obstime is the time of observation, and patient is the patient id. My linear mixed ...
0
votes
0answers
16 views

Multiple Regression, good P-value, but Low R2

I am trying to build a model in R to predict Conversion Rate (CR) based on age, gender, and interest (and also the campaign_Id): The CR values look like this: The correlation coefficients are not ...
0
votes
0answers
7 views

Are the changing of correlations between features and target variables over time a big reason for the failure of predictive models?

Would love to know your thoughts on what I believe could be an important issue in predictive modelling! At the moment I'm really interesting and curious about what makes a predictive model fail. In ...
1
vote
2answers
34 views

Timestamps in Ridge Regression Scikit Learn

I am trying to transform data for use in regression, most likely the Ridge or Lasso technique implemented in sklearn.linear_model. My training data contains time ...
1
vote
0answers
11 views

Testing a Multiple Logistic regression Model Goodness of Fit and accuracy

I am trying to create a predictive model that will determine the likelihood that a lead will take a specific action. I used a Logistic regression because either the Lead does (1) or does not (0) take ...
0
votes
1answer
22 views

What to do first: Model Selection or Predictor selection?

so I am trying to look at a NBA player data set and create some predictions based on regression analysis. But I am stuck wondering what steps do I do first. Should I find the model I want by checking ...
0
votes
0answers
26 views

Alternatives to Neuronal Nets and Gradient Boosting for undifferentiable objectives

It may occur that one has to solve a ML problem but wants to achieve the best result w.r.t a metric that may not be differentiated. This directly implies that such a metric may not be passed to ...
0
votes
0answers
26 views

Choosing a model

I am working on a sales forecast right now and I have created 4 models but I am unsure which one to use. I have 17 Quarters of data(4 Full years + 1 QTR) and I am only looking to forecast 2 quarters ...
0
votes
0answers
22 views

Why is comparing models based on likelihood acceptable?

Suppose for instance I have a set of models and I want to choose the best one. One way I've seen is to use something like AIC or BIC. I'm not sure how higher likelihood translates into a better model -...
0
votes
1answer
37 views

How to decide moving window size for time series prediction?

I have a model to predict +1 day ahead of this time series. Looking at the chart you can notice some seasonality every 5 days. I suspect using a moving window as training set could help me making a ...
0
votes
0answers
12 views

low prediction accuracy with unbalanced Dataset

I am working on a binary classification problem, where I have 2 classes (0 and 1). I have created a balanced Dataset with 70k samples(50% have the class 0 and the other 50 % the class 1), trained a ...
1
vote
0answers
14 views

How can I add interaction term or transform the variable in randomforest? [duplicate]

I am making predictive model with randomforest in R's randomForest package. I wonder how determine independent variable's transformation and adding interaction term. In regression model, the reason ...
1
vote
0answers
20 views

Should I lag explanatory variables in regression with apparently strong predictive relationship?

I'm no expert when it comes to statistics (learning though) and I am working on developing a multiple linear regression model in an attempt to forecast sales revenue. I feel like I may have developed ...
1
vote
0answers
27 views

How to generate predictive values when using Anova (Type III) on mixed effects linear model

I am running a mixed effects linear model on AB test and using Anova, Type III to determine significance. I am doing this because I have an interaction variable. My R code is : ...
9
votes
2answers
215 views

A ''significant variable'' that does not improve out-of-sample predictions - how to interpret?

I have a question that I think will be quite basic to a lot of users. Im using linear regression models to (i) investigate the relationship of several explanatory variables and my response variable ...
1
vote
0answers
26 views

How to apply a model built using Multiple Imputation to predict on dataset with missing data?

I understand that Professor Harrell recommends using the target variable in Multiple Imputation. An example using aregImpute of the rms package is in his lecture notes: http://hbiostat.org/doc/rms....
1
vote
1answer
26 views

Can Negative Predictive Value and Positive Predictive Values be the same?

Is there any scenario where a negative predictive value and a positive predictive value would be the same? Specifically, when using a neural network for binary prediction. Can this be a sign of ...
0
votes
0answers
23 views

Why does my precision, recall, f1score and accuracy decreases when I am feeding the model with Upsampled dataset?

I am currently working on a dataset which is imbalanced. It has 2850 negative label points and 483 positive label points and I do upsample on the dataset to balance it. But my model performance ...
0
votes
1answer
27 views

How do you build and train a model using time-series data?

Pardon the 101 level question. If I need to read up on some basics, please point me to few good resources. Let's say we want to predict the Blood Sugar Level at the next 60 minute instant given a ...
0
votes
0answers
10 views

Calculating probability of churn in year 2 and year 3

I am trying to find customer future value. I am using a very simple formula here. cfv for year 1 = yearly gross margin or profit * probability of retention. Now if I want to find cfv for year 2, then ...
0
votes
0answers
20 views

Estimating a smooth probability distribution

I have a data frame storing labels and numerical values, e.g. cars and measured speed ...
0
votes
0answers
7 views

Statistical evaluation of a diagnostic model versus a prognostic model

I aim to compare the performance of a diagnostic model (with a binary outcome) to a prognostic model (survival). The diagnostic model was created using a random forest algorithm after univariate ...
1
vote
0answers
16 views

At a loss regarding feature selection vs coefficient estimation. Can you ever re-do the latter after the former?

I'm looking at a binary classification problem where p>>>n (9,000 gene expression variables for 290 patients who either have or don't have disease). I hypothesized that it would be easy to find "...
1
vote
2answers
104 views

Why is linear regression overestimating small values and underestimating big values?

I am trying to predict age from a couple of variables using linear regression, but when I plot predicted age against real age, I can see that small values are significantly overestimated and big ...
0
votes
0answers
13 views

checking bi-variate relations in predictive model

Friends, I am using decision tree and logistic regression for prediction purposes (my dependent variable is a binary variable). I am just wondering whether I need to check chi-square (for categorical ...
0
votes
1answer
17 views

Manga translation updates: what kind of data/what model?

Long story short, I'm trying to predict how likely it is for a content creator to release new content or when they are most likely to do so (and possibly how this changes over time). My problem is ...
1
vote
1answer
166 views

Model for predicting chance of winning in variable count of opponents

I have dataset with horse racing results including bookie odds - converted to percentage chance of winning. Data are stored in relation tables. The basic entity relation is described on image. Each ...
1
vote
0answers
28 views

Probability distribution model [closed]

I have a dataset in the following form: ...
1
vote
1answer
22 views

Use Logistic Regression to Predict Click Through Rate

The goal is to predict click through rate of article content. Currently, the linear regression is used and the input data set is at article level. The label is the click rate of the article. The issue ...
1
vote
0answers
11 views

Using the Volume of Sales obtained today to predict volume of Sales at the Event

I wonder if anyone can help. I have a set of data on event ticket sales. I have information on eventdate, location, capacity, cumulative sales, sales date, total sales. I want to be able to build a ...
0
votes
0answers
17 views

How many time series is it logical to model using VAR (in R)

I have weekly sales for around 6000 products, for which I would like to obtain forecasts for n periods. I believe that estimate ...
1
vote
0answers
37 views

How people use Stacking method in the real-world problems? [closed]

I know that stacking is a very strong method when you using it in machine learning competition(Like Kaggle). But in real life situations do people use this for modeling very often? I heard that ...
1
vote
1answer
41 views

How to check if a polynomial regression has any predictive value? [closed]

After fitting the polynomial data to a given curve, how can I check which of the many curves has the most predictive value ?
1
vote
0answers
10 views

How to minimize dependency between existing model and retraining dataset (using fraud modeling as example)?

What are methods to minimize dependencies between the existing model in production and the data that will be used for retraining future models (when future samples are created based on the existing ...
2
votes
1answer
41 views

When to apply time series models?

When should I apply traditional time series models (e.g. additive, ARIMA) versus other models such as linear regression? For example, if I wanted to build a predictive model for website traffic, I ...
1
vote
0answers
24 views

variable transformation

friends, I want to use regression for prediction purposes with a sample of over 20k. I have the following results for the OLS with adj R-squared of 0.64: my first question is: can I say that the ...
0
votes
1answer
31 views

What to do when the test data set has many “features” that are generated by dummfying a categorical variable that are not present in the training set

Say you have a variable (in this case industry) that you dummify (one hot encode) hence creating many new features in both the training and test sets for which you are getting ready to run a machine ...
0
votes
1answer
15 views

Leveraging Images in Random Forest Predictive Model

I am using a random forest to make numerical predictions for the performance of products using structured variables, and am looking to leverage images to improve my predictions. One idea I have is to ...
6
votes
2answers
184 views

Is it wise to use predicted values to model predicted values further down the line?

Hi I have two questions that are related. I am wanting to model sales for different areas in a business and have been looking at ARIMA, I am not too happy with the results of this especially when I ...
1
vote
0answers
30 views

Multivariate Time Series with Long Horizon Prediction [duplicate]

There must be a model for what I would like to do, because I have been spending far too long with this project for an answer not to exist. For reference, I am using R. I have a time series object, $...
2
votes
2answers
123 views

ARIMA Model configuration for hourly forecasting problem

I have a database based on hourly data and I need to forecast next 24h of a single variable. I was thinking about applying an ARIMA model with some exogenous variables but I don't succeed to configure ...
0
votes
0answers
13 views

What is the appropriate model for data with a 1:n response to independent variable ratio

I suspect I am at a loss of terminology, and will accept "Google search terms [x, y, z]" as an acceptable answer in this case. The nature of the process being observed is as follows: a certain Goo ...
1
vote
0answers
27 views

How to estimate the standard error of the leave-one-out cross-validation estimate of the prediction error?

How does one estimate the standard error of the leave-one-out cross-validation estimate of the prediction error? For each fold (leave out the $i^{th}$ observation), the LOOCV estimate of the ...
2
votes
1answer
30 views

Choice of hyper-parameters for Recursive Feature Elimination (SVM)

Suppose that I am interested in applying Recursive Feature Elimination (such as RFE() in sklearn), and my data set is of medium ...
0
votes
0answers
10 views

LSTM for extreme event prediction: scaling factor between observation X and predicted time serie

I have an observation X every hour => Time serie 1. I want to predict an event Y such as Y =f(X) => Time serie 2 : Y_pred. For the past periods I have both X and Y (Y_true) For future periods I have X ...
0
votes
0answers
16 views

Comparing overall and conditional models

Let's say I have two regression models with 5 variables, one that is for all observations, and one that only includes values when the temperature is higher than 75 degrees. As an example I have this: ...
-1
votes
1answer
29 views

Using machine learning model for predictions

I am trying to use my random forest model for predictions. I only want to select important variables(ex:top 50), and use the saved RF model to predict the response variable, changing the predictor ...