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 ...

learn more… | top users | synonyms

0
votes
1answer
22 views

Applying randomforest algorithm (fit) on new data without recomputing the fit [on hold]

I have a need to do realtime predictions for individual rows of data based on a previously computed randomForest algorithm. How can I run the "predict" command without recomputing "fit" on the entire ...
1
vote
0answers
16 views

Sample size needed to validate classification/prediction model

Dose any rule of thumb exist (or possible calculation) regarding sample size needed to validate an binary classification model. We have developed this prediction model for a medical condition and ...
0
votes
0answers
22 views

Why Binning Variables in Predictive Analytics?

Lot of discussion in CrossValidated focuses on optimal binning methods, binning example etc. But I am trying to figure out what are the scenarios that I have to bin variables whereas it's better idea ...
1
vote
0answers
13 views

Prediction model on hybrid data

I am currently working with a data set where I have both continuous, discrete and categorical (without any order) data. And I have to predict a continuous data. To be concrete, my problem is a ...
2
votes
2answers
45 views

Combination Forecast - Which models to pick?

Combination Forecasting can be produced by simply averaging different forecasts or employing more complex techniques (see Makridakis, 1989; De Gooijer and Hyndman, 2006; Goodwin, 2009; Pesaran and ...
1
vote
0answers
17 views

How to merge different predictive models training with different data sets?

Is there any good method to merge/consolidation different predictive models which were trained on different features but outputs the same goal. Example: Model 1 with features a + b + c (trained on ...
4
votes
1answer
31 views

How to train a model when instead of a target we have a range where it is?

Often in machine learning we have a situation when target is numeric (real or integer). Each target comes with an associated input vector. The goal is to learn the mapping from the input vectors to ...
3
votes
1answer
29 views

How to make a trained neural network “forget” an instance?

I am using neural networks for predicting the behavior of a dynamic system. A neural network is trained online using snapshots from the system's past. The system changes its state at irregular ...
3
votes
1answer
35 views

Weighted sampling as a way to eliminate specific source of variation?

I am facing a problem of predicting probability of an event given two correlated predictors where only one of them is of interest. Thus, I’m trying to eliminate one of them from the model while making ...
1
vote
0answers
31 views

Best approach to predict significant factors without any complete cases

I have a dataset that contains records of donors with various biographical info (city, state, zip, number of children) and the total amount they donated over 10 years. Some never donated and thus the ...
0
votes
0answers
5 views

Penalize a long-form panel linear regression prediction?

What is the recommended penalty, if any, for a long-form panel when calculating multiple linear regression $\hat\beta$ parameters or predicting single responses from (unobserved) values of $X_i$ in ...
1
vote
1answer
54 views

Explain “validation” process of repeated k-fold cross-validation?

My understanding is currently that the canonical repeated k-fold cross-validation (CV) process might do the following if $n=100$ observations in sample, $k=5$ folds, $i= 10$ iterations (see iteration ...
0
votes
1answer
15 views

Estimate linear regression paramaters with chain modeling for longitudinal data?

Within a frequentist, deterministic paradigm of multiple linear regression, is there a (standard) method to accomplish "chain modeling for panel data" in a way that avoids formal identity (and/or ...
0
votes
1answer
23 views

Change settings in the prediction model (caret package)

I am using the package caret and GBM method for my predictions. ...
0
votes
0answers
13 views

Which distribution should I better use to predict the response in {0,20} applying GBM? [duplicate]

I want to predict the response that is in {0,20}. I am using GBM to make the prediction. ...
0
votes
0answers
12 views

scoring/predicting for new observations

I have two data sets of variables where one of them - the new observations - has no dependent variable. The data set without a dependent variable has around 20 times the number of records. ...
0
votes
0answers
18 views

Can auto-predicted values ever improve linear regression?

You want to predict values of $y$ using a linear model of the following form: $ y = \beta_0 + \beta_1x_1 + \beta_2x_2 + \beta_3x_3$ $y$ is significantly dependent upon all three variables. ...
0
votes
0answers
16 views

On the prediction mean square error of a model

Suppose my model is $y_t = \alpha + \beta t + \epsilon_t$ the l-step-ahead prediction is given by $\hat{y}_{T+l | T} = a + b(T + l)$ where $a$ and $b$ are the OLS estimators of $\alpha$ and $\beta$. ...
0
votes
0answers
33 views

Should significance of each parameter be proved?

Suppose I want to prove the following linear regression, which represents a predictive model of $I$: $$ I = \beta_a A + \beta_p P + \beta_d D + \beta_s S + \varepsilon $$ Here $\beta$ are regression ...
6
votes
1answer
151 views

What is shrinkage?

The word shrinkage gets thrown around a lot in certain circles. But what is shrinkage, there does not seem to be a clear definition. If I have a time series (or any collection of observations of some ...
4
votes
0answers
16 views

R HoltWinters - Odd dip in predictions

So, I have this time series that tracks the daily number of applications to a graduate program. Each application period is 64 days - so for each period, you start at zero and it goes up until the ...
0
votes
0answers
33 views

using biomod2 package with continuous response variables

I am looking to correlate crop area with climate variables and then predict for future if suitability of crop areas will change/remain same under different climate scenarios (in line with species ...
1
vote
0answers
37 views

Discrimination between measurements made at different points in time

I would like to ascertain what variables discriminate best between experimental conditions in a repeated-measures experimental design. I have performed Repeated Measures MANOVA to determine whether ...
2
votes
1answer
77 views

How to score predictions in test set taking into account the full predictive posterior distribution?

I have three predictive models (regressions) which parameters are estimated by Markov Chain Monte Carlo. Predictions are made over a test set of size $N$. Since I compare the models under different ...
1
vote
0answers
43 views

Incorporating intraday data into end-of-day forecast

my target variable is observable intraday but I am interested only in EOD forecasts. I will denote the variable $\ y_{D,24}$ as the reading of interest for day D is ...
3
votes
0answers
42 views

Modelling flight delays with negative values

Modelling flight delays with negative values I am working on a model to predict whether a flight will be delayed. The data consists of some explanatory variables for flights from a specific airport. ...
0
votes
0answers
10 views

Comparing the impact of 2 independent variables on the dependent

I'm using a predictive modelling technique which has 2 parameters. I've performed a sweep of values for each of these 2 parameters, running each permutation of parameters 30 times as the technique is ...
0
votes
0answers
24 views

Incorporating systematic error in (spatial) predictive modelling

I have created a model (random forest) and withheld 20%. When I apply the model to the withheld dataset and check the residuals against the real values I can see there is a systematic error e.g lower ...
0
votes
0answers
26 views

How can I use the output of KODAMA to predict unknown data points?

I can use KODAMA to create a model that classifies input data into two groups by setting the W vector to indicate the group and fix to a vector of all ...
0
votes
1answer
25 views

One-Step ahead predictive likelihood for time series forecasting

I am still new to Bayesian forecasting, so I am hoping to get some clarification on a simple concept (by the sounds of it). Suppose that we are interested in forecasting some time series one-step ...
0
votes
1answer
16 views

Predicting Arrival/Departure of butterflies

I don't have a solid background in statistics. I am double checking with you on a phenomenon I am trying to study. we are doing a study of some very rare species of flowers. We are putting them in ...
0
votes
1answer
17 views

train multiples observations from the same person in caret

I have data where persons were give four different tasks under three different conditions (intensities). The data looks like this: ...
1
vote
1answer
31 views

What predictive models allow me to make new predictions on a series in constant time, without needing to recompute previous ones?

I'm a software developer working on a system that stores thousands of independent metrics, each with several tens of thousands of timeseries data points. We'd like to make predictions about where the ...
0
votes
0answers
63 views

Overfitted Cox regression

I'm trying to compare the prediction abilities for death of two new biomarkers using a cohort of 173 patients. My problem is that I have only 31 outcomes and my baseline model (the one without any ...
0
votes
0answers
10 views

What models allow the study of the relation between a set of response variables and a set of covariates?

A first technique that comes to mind is Canonical Correlation Analysis. Bayesian Networks and other graphical models, I guess, can also be used to analyse such things. Any else that I should be aware ...
0
votes
1answer
58 views

Choosing a good binary classifier to be trained by a small set of labeled data

I have a small set of labeled data (diagnosis in individual subjects): ~50 of "sick" observations ~100 of "healthy" observations In reality, only ~1% of the observations are expected to be ...
6
votes
3answers
232 views

Deciding between a linear regression model or non-linear regression model

How should one decide between using a linear regression model or non-linear regression model? My goal is to predict Y. In case of simple $x$ and $y$ dataset I could easily decide which regression ...
1
vote
1answer
47 views

Conditional distribution for Exponential family

We have a random variable $X$ that belongs to the exponential family with p.d.f. $$ P_X(x|\boldsymbol \theta) = h(x) \exp\left(\eta({\boldsymbol \theta}) . T(x) - A({\boldsymbol \theta}) \right) $$ ...
0
votes
0answers
17 views

How to predict values? [duplicate]

I have a simple time series with at least a measurement a day. I would like to know if there are algorithms that can deal with missing values and measurements that are not taken always at the same ...
0
votes
0answers
12 views

Apply Cross Validation to a prediction linear regression model

Hello I am slightly confused on how to apply a k-fold cross validation to a prediction linear regression model. Also how can i look at the data that the prediction model outputs before i apply cross ...
0
votes
2answers
49 views

Predictive Model - Increase Pediction Accuracy for Less Likely Events

I am trying to build a model that predicts the which binary category a respondent belongs to (0 or 1). I have demographic variables (all categorical) and a few 10 point questions. I have built a few ...
0
votes
0answers
37 views

Best prediction model for binary data

I have a large medical data set (100.000+ patients), and have many variables (but selected 20 most interesting ones - 15 of these are binary). I want to make a model that predict if these patients ...
0
votes
2answers
59 views

When make clusters in a predictive glm model?

If I want to build a predictive glm model, should I make cluster analysis on 100% of observations or on training sample (80%)? Thanks
3
votes
2answers
43 views

What is the proper name of a model that takes as input the output of another model?

Thanks in advance for the help. I am writing a paper and for the life of me can't remember the proper term for a model that works as follows. ...
0
votes
1answer
54 views

Questions about weather prediction in scikit learn

Hello I am a high school student doing research on weather. I have a dataset that has four columns each labeled with time, pressure, and lat/long. I am confused on the cross validation process. What ...
1
vote
1answer
45 views

Is it better to use MAE or MSE for perfomance measure? [duplicate]

My data set is about forest fires in Portugal. I want to define a model that can predict better wildfires. In my data set, the outliers are entries referring to big fires. What is the best performance ...
0
votes
0answers
13 views
3
votes
2answers
100 views

predictive modeling: comparing actual and predicted values in terms of accuracy

I have applied a predictive model on a hold out data set on which I know actual values of the target variable. I wonder how to compare actual and predicted forecasted values, verifying whether on ...
2
votes
1answer
44 views

What does it actually mean for classes to be balanced?

I saw the following statement when reading Kuhn's APM: "The classes are fairly balanced; there are 111 samples in the first class and 97 in the second..." I thought balance would require the ...
9
votes
1answer
503 views

How fair is it to use the word “predict” for (logistic) regression?

My understanding is that even regression does not give causality. It can only give association between y variable and x variables and possibly a direction. Am I correct? I've often found phrases ...