Stack Exchange Network

Stack Exchange network consists of 175 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

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.

0
votes
0answers
9 views

KNN works in `class` but not `caret` (Too many ties) [on hold]

I am making a KNN algorithm to predict close_price with about 80,000 rows of this data. ...
0
votes
0answers
16 views

What does the intercept represent in a model matrix?

I am making a KNN algorithm to predict close_price with about 80,000 rows of this data. ...
0
votes
0answers
13 views

Predict response and replace values not in original fit using the mean [on hold]

I have fitted a model based on a subsample of a bigger data set. However, when I run the command to predict values of the whole dataset, I am getting the following error: ...
0
votes
0answers
12 views

Variant of validation with singleton test sets

Is the following approach to model validation somehow reasonable? And is there a name for that approach? We have 110 data points, iid assumption holds and we want to compare two predictive models M1 ...
0
votes
0answers
5 views

Methods to predict a binary outcome with ordinal covariates

Just a very general question: what are the most efficient methods to predict one single binary variable $Y$ using a set of ordinal covariates? In my precise case, I have around twenty ordinal ...
0
votes
0answers
10 views

Quick question on the stationarity of an autoregressive process depending on the time index

Let the stochastic process $\{ X_t \}_{t \in \mathbb{Z}}$ satisfy the equation $$ X_t = \theta X_{t-1} +\epsilon_t $$ where $|\theta| < 1 $ and $\epsilon_t$ is gaussian white noise. This is an ...
0
votes
0answers
15 views

Effect of the presence and absence of predictors to the classification outcome

I have trained a classification RF model and then ranked the predictors based on their contribution (importance) to the model. How can I estimate the effect of each individual predictor in the ...
1
vote
0answers
16 views

Random Forrest data prediction (in R) and data bias

I have a data set from 2014 to the present and I am trying to classify stock performance based on whether the stock outperforms the market by 5% or not (1 vs 0). I am using a random forest model in R (...
0
votes
1answer
16 views

Variables preparation for MLP regression

I'm trying to create a regression model with MLP to predict a continuous variable, that is the income of a movie. My set of regressors is composed by around 15 binary variables (I've used one-hot-...
0
votes
0answers
15 views

Does low predicted variable significance indicate a model problem?

Suppose I am trying to construct a temporal model of some variable of interest. Suppose I have the following situation, where voi is a variable of interest, t is time, and x is some other variable: <...
0
votes
1answer
19 views

Prediction of a continuous variable

I'm trying to create a model to predict a continuous variable, that is the revenue of a movie, given many predictors, such as its budget, the length of the film, the genre... I'm planning to use MLP, ...
5
votes
1answer
231 views

Model for spatiotemporal and discrete variables

I have a situation where I am monitoring events at 50 or so geographical sites in a town and at each of these sites, I am making measurements regarding the count of certain particles (so the ...
0
votes
0answers
7 views

Query Tuning RF and Neural network model output

I am working on UCI wine quality data set(total 5k records).Below is the small snap of the data.I have to predict quality(last col) which has scale of 1-10. I am trying to see the accuracy- using ...
0
votes
0answers
26 views

Maximum R2 of predictions based on correlation

Is there a way that one can tell the maximum R2 or prediction algorithms can be achieved depending on the correlations between variables or using another way? And does this have to do with Rademacher ...
0
votes
1answer
20 views

How to determine N of LOOCV

In my textbook, it says that LOOCV is where $K=N$, but how do I find the value of $N$? Is it just $K-1$?
0
votes
1answer
27 views

Precision/recall curve and accuracy/threshold interpretation

I am running a logisitc regression and trying to interpret the predictive power it generates. How should I interpret the precision/recall curve and the accuracy as a function of threshold? My ...
0
votes
0answers
17 views

Formal Definition of Over-Predictive Model

I am looking for a formal definition or criterion to determine whether or not a model is over predictive. My understanding of a model being over-predictive, is a parametric model whose parameters are ...
0
votes
0answers
8 views

notation of derivative probability function

I am reading article about bayesian predictive function. In the article it denote posterior distribution $\pi_n(d\theta) = \frac {\prod^n_{i=1}f(y_i|\theta) \pi(d\theta)}{\int \prod^n_{i=1}f(y_i|\...
3
votes
0answers
17 views

How to combine noisy and noise-free datasets to train a model

Overview Suppose I have two datasets, both of which consist of rows of features and their matching labels. One of these datasets is noise-free and its labels correspond to the ground truth, but the ...
2
votes
0answers
93 views

Leave-one-out cross validation in regression: R squared cannot be used - how else may model performance on unseen data be evaluated?

I have a regression problem with very few datapoints, therefore I want to use leave-one-out cross validation (effectively N-fold cross validation with N being the number of datapoints) to determine ...
0
votes
0answers
93 views

How to automate time series analysis/monitor (descriptive)?

For example, here is a traffic graph from google analytics: You can see that first there is a spike, and then a long plateau time, and then increase to a slightly high stage (but more volatile), and ...
0
votes
0answers
29 views

Compute predictive power variable on stock returns

I have a self made dataset with 40 stocks listed with following information in a range for 1 month: - closing price stock - sentiment score e.g. 01-01-2018 - closing price - sentiment score for that ...
0
votes
0answers
7 views

Testing Response Model

I'm thinking about strategies to examine how well a churn model performs. And, I know one question I'll receive is can we just target customers in the top decile of the churn model to evaluate model ...
0
votes
0answers
18 views

How to featurize over a moving window of Time series data?

lets say we need to predict future values(like taxi demand) of a time series data(like pickups over a period of time from past to present) To build a model I created 'n' data points at each point of ...
3
votes
2answers
75 views

Does zero-one loss correspond to any maximum likelihood procedure? [duplicate]

Squared loss for linear regression corresponds to a MLE of a Gaussian model, and cross entropy loss corresponds to MLE of a logistic model with discrete probabilities. Can zero-one loss be ...
0
votes
0answers
20 views

Calculating predicted value from polynomial regression by hand

I have the following output from from R. I know we can calculate predicted values using the predict() function. How can we calculate the predicted value by hand if X=2 or 3? I'm using the below ...
0
votes
0answers
27 views

Logistic Regression - Coefficients not defined because of singularities

I am running a regression model to predict dropout from an online program. People have to take 5 classes but some people dropped before taking the 5 courses. So I am using a dummy variables that is 1 ...
0
votes
1answer
113 views

Compute standard deviation of accuracy

edit - more information about what the code given should represent The following pseudocode outlines the problem as I have it ...
0
votes
0answers
18 views

Response variable is right skewed

I'm doing a project where in I came across my response variable to be rightly skewed. If I run the linear regression model or MARS it gives me Rsq as 1. I don't know why is it showing and also doesn't ...
1
vote
0answers
13 views

Predict value based on independent values and time [closed]

I want to use Python to predict a value of a chemical reaction. As an input I have time units (0,2,4..) and the concentrations of 2 solutions. As an output I have a chemical measurement. As an ...
0
votes
0answers
17 views

Infer statistical model prediction confidence coefficient based on the past predicted and actual results

Currently, I encountered such as a problem: For example, when I built a statistical model based on the past 3 month data and using the model to predict for the next month. But I found that the ...
0
votes
0answers
18 views

Location-based time series: predicting Uber surge pricing

As a learning exercise, I'm working on putting together a few different machine learning projects that I can spend some time honing and studying. One of these projects is predicting the locations of ...
1
vote
2answers
33 views

What is the use of knowing the underlying data distribution?

I am trying to understand few basic concepts of data mining, machine learning, etc. and I am new to this field Say that I have a data sample and I have done maximum likelihood estimation or some ...
0
votes
1answer
19 views

How do I estimate probability of success with no successes? [duplicate]

My $6$ friends and I tried buying tickets to a popular event. Everyone who wanted a ticket got a random number and if your number is less than or equal the number of tickets available, you can buy a ...
1
vote
0answers
54 views

GLM Frequency and Severity Models. How Do I improve from here? (R code) [closed]

Background: I've been tasked with creating a rating model by Peril using GLMs. It's commercial lines property, so the data is pretty sparse. The carriers have been asking for Premiums by peril, so we'...
1
vote
1answer
54 views

Monthly Times Series Modeling Approach

I have a machine learning problem and have been working in Sklearn/Pandas with Python to come up with an accurate model. I find myself deep in a rabbit hole trying to learn the best approach and how ...
1
vote
1answer
24 views

Using lagged explanatory variables to forecast future value of depended

Is there a way or method to use older values (lagged) of independent variables with alternative lags to explain current value of dependent variable? For time series specific
1
vote
1answer
27 views

How to extrapolate partial curve based on other complete curves?

I heat a room to a certain temperature. Then I let it cool over time. I measure the temperature at 6 intervals as it cools. I repeat this process for some other rooms. From this, I get a set of ...
3
votes
1answer
33 views

Joint AR(1) posterior distribution explicit under conjugate prior

I have encountered a problem in my textbook 'The Bayesian Choice' by Christian P. Robert. It goes something like this: $"$For a particular case of AR(1) model, $(x_t)_{1\leq t\leq T}$. Where $x_t = \...
0
votes
0answers
35 views

Model Validation Through Bootstrapping Optimism; p >>> n

On this forum, I've read quite a few posts on the use of bootstrapping a statistic known as optimism when evaluating various models for their out of sample, predictive performance. I personally have ...
2
votes
1answer
36 views

Out-of-sample predictions for mixed model are the same as naive model (ignoring the random effects)

I have a dataset that consists of subjects coming into the clinic (for treatment of another disease) and they are screened for Tuberclosis (as they are a high risk population). Every time they are ...
1
vote
1answer
38 views

Bayesian approach to interval prediction?

I am working on a problem involving understanding/predicting customer frequency. The data I am working with is structured as a series of interval days between orders: ...
1
vote
1answer
41 views

Brier Score of a Prediction, Mathematical Notation

I am currently working on a logistic regression model that is fitted on the base of a training set ($D_0$) and is used to predict the outcome (0 or 1) of an independent test set ($D_1$). As an ...
0
votes
0answers
22 views

Multiple Imputation Vs Pool

I have simple question after running multiple imputation what the purpose of pooling? Suppose if i run a multiple imputation using method cart , after running this imputation technique i get very ...
11
votes
2answers
1k views

In Bayesian inference, why are some terms dropped from the posterior predictive?

In Kevin Murphy's Conjugate Bayesian analysis of the Gaussian distribution, he writes that the posterior predictive distribution is $$ p(x \mid D) = \int p(x \mid \theta) p(\theta \mid D) d \theta $$ ...
0
votes
0answers
14 views

Prospective Statistical Techniques for Clinical Data

Can anyone point me to a beginner-friendly, less math intensive resource for studying clinical data prospectively? In a lot of my data sets, I only have a baseline measurement. I have repeat measures ...
3
votes
1answer
231 views

Does regularization penalize models that are simpler than needed?

Yes, regularization penalizes models that are more complex than needed. But does it also penalize models that are simpler than needed?
0
votes
0answers
44 views

Multi State Models to analyze/plot disease progression and probability of being misdiagnosed

Let's say that I have the following dataset containing information for 100 patients that have been followed up for a certain number of years to check if they develop a certain disease. We know up-...
0
votes
1answer
26 views

Statistical prediction model with copula

Copula models are used widely to present the dependency structures among variables. Assume that I have a disease dataset. Assume further that I need to diagnose patients. Suppose that ...