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
0answers
17 views

Better Predictive Model?

I'm running a bunch of different models trying to find one that is best at predicting using a Validation set and Root Average Squared Error(RASE) calculated from residuals as my main criteria. My data ...
-2
votes
0answers
20 views

Predictive modelling problem

I have a data set on Effort Deviation of software projects. The data is as follows: 985 data points. Response is continuous with negative, positive, and zero values. Input Parameters are 15 in ...
2
votes
0answers
8 views

dividing a multiply imputed dataset into derivation and validation cohorts

R/statistics noob. Mac OSX 10.11, RStudio 0.99.842. I'm developing a clinical prediction tool as part of my PhD. I have missing data (23k cases, 24 variables, 70% of variables have at least one ...
2
votes
0answers
15 views

Summing estimated probabilities in classification model

I am developing a series of predictive classification models (decision trees, neural networks, etc.) aimed at predicting the number of people who enroll in a program based on a variety of demographic ...
0
votes
1answer
12 views

Why do gam predictions not match gam smooth?

I am studying the effect of organic farming on honey reserves in honeybee colonies. I am trying to see how an increase in the percentage of organically farmed land (at various buffers around bee ...
0
votes
0answers
8 views

Difference between weighting and replicating observations in linear regression

I have a model in which each case is summary statistic of many observations. I am using a mixed effect model (lmer() in R) for prediction and I thought to give ...
0
votes
1answer
11 views

taking average of several models and feature sets

just a quick question that i cant seem to find a definitive answer for. When im doing feature selection, i end up with a list of the top performing sets. Would it make sense to use the top 10 sets ...
0
votes
0answers
28 views

Stuck while trying to predict on data based on H2O Deep Learning model

I have created an H2O Deep Learning model in R for multi-class classification and I want to use it to perform prediction. I would have assumed that if I use the model to predict on the validation ...
0
votes
0answers
11 views

Adversarial sequential learning with a linear model

I have a problem with the following characteristics: The value of an observation is a function of its predictors The nature of the relationship between value and predictors changes slowly over time ...
1
vote
0answers
12 views

Predicting occurrence of an event using time series data

I have data coming from sensors for 1 month. The data is time series with each data point separated by interval of 1 second. There are predictors like temperature, pressure, speed of the fan that has ...
0
votes
0answers
14 views

Method to determine whether or not users had a bad experience based on multiple variables: Average Bandwidth, Latency, and frame rate

I would like a recommendation on the best statistical method to use, as well as any suggested R packages to achieve this goal. I have three variables, Bandwidth, Latency, and frame rate for a set of ...
0
votes
0answers
7 views

A rate model for sodium channels

I am studying by myself Human Physiology. I have encountered the following question: In the following given model of sodium channel with 3 states open closed blocked (which I assume means ...
3
votes
1answer
26 views

Single prediction vs. summing more granular n-step ahead predictions

Say I want to predict the total rainfall for the next 365 days based on a set of predictors and daily historical observations. I could build a model that predicts annual rainfall and make a single ...
1
vote
1answer
24 views

With multinomial regression, how to predict an event and get the ROC curve?

I'm using the multinom package in R to run a multinomial logistic regression model. My dependent variable has 3 levels and as the output, I'm getting the ...
0
votes
0answers
11 views

predict a distribution of efforts over time

I am given data with efforts (the mythical man months) for many (ca 350) projects over 2 years. We want to use this data to predict the evolution of upcoming projects in terms of staff to find out ...
0
votes
0answers
11 views

adjusting nnet model for prediction

Could someone give some hints how to adjust paramters in nnet model for predictions ? I mean following parameters: maxit, range, decay, size, and ...
0
votes
0answers
11 views

How to get correlation of each predictor to response

I am wondering how can I get the correlation from one predictor to a repsonse when I am looking at a given data set with many predictors. For example, the output of GLM in R would be exactly what I ...
1
vote
0answers
10 views

Cross-classified multi-level model - application to marketing

I am working on predicting whether an individual customer will respond favourably to a marketing campaign (yes/no). I have data about customers, and their responses to previous campaigns. If possible, ...
0
votes
1answer
51 views

What model to use for predicting future expenses of an individual?

I am currently working on a Personal Finance application, which tracks expenses of a person. When entering an expense entry, the user selects the category of the transaction (e.g. 'Bills', 'Food', ...
0
votes
1answer
29 views

Can we use network neural for nonumeric data?

I am going to use network net package to do predidction. Tell me please: Is it neccessary to do scaling of data ? Is it neccessary to convert all data into numeric/int type ? I am newbie at this ...
0
votes
0answers
24 views

How to build a model from data with a proper hypothesis

I have a large dataset of items in a store and how they sell. It looks somewhat like this: ...
0
votes
0answers
9 views

Polynomial curve fitting for temperature prediction

First of all, I would like to say that I know very little about statistics. I need to make a C# application to predict three days weather for school project and need some model and have been exploring ...
0
votes
1answer
5 views

Classification of temporal instances

Problem description I am working for a telecom domain project where I am tasked to predict whether a customer would dispute his/her monthly bill or not. I have following historical data elements ...
5
votes
1answer
77 views

Reproducible benchmarks for the performance of statistical prediction methods?

There are many statistical models used for predictive modelling. These include famous methods such as naive Bayes, knn, SVM, random forest etc. I am looking for reproducible examples (preferably in ...
1
vote
1answer
31 views

Machine learning step order question

I have been working on this project for over a year now and I believe i finally have things figured out. Mainly i'm looking for any suggestions or things i'm doing wrong with my process, but i also ...
0
votes
0answers
17 views

multi linear regression model. Modeling with a heavily skewed binary independent variable

Dataset and goal: One continuous measurement( to be modeled as a dependent variable) and four other measurements (one binary and the rest are category variables with multiple levels) to be modeled as ...
0
votes
1answer
59 views

Noob Data Design Question - 1 on 1 competition

My question is about how to best structure my dataset for a competition between 2 players of a game (for the purposes of prediction of future game winners). There will potentially be hundreds of ...
0
votes
1answer
11 views

Conditional probability sub-model so solve setting with a factor that has many levels

I stumbled upon a post of the http://www.win-vector.com/ blog where they treat the problem when a factor with many levels occurs. In my understanding instead of using the factor itself, they use the ...
0
votes
0answers
10 views

Topics needed to learn to for predicting future of a data set?

I am planning to work this summer by writing a code which will eventually become an app, that has the ability to predict what the prices will be atleast 6 month ahead of time. I have a data set of the ...
1
vote
0answers
30 views

Organic vs Paid Attribution Model (Granger)

I'm wondering if there is literature or studies done on how to model organic attribution from paid user acquisition. So the context is, on our mobile app, we have paid installs that we purchase and ...
0
votes
1answer
36 views

ARMA lag selection for ARMA-GARCH models

When I read this group questions about lag selection for ARMA part of ARMA-GARCH models I found 2 different answers from moderator: The use of GARCH and ARMA GARCH estimation process in practice I ...
0
votes
0answers
18 views

Time Series Modelling With Two (or more) Periodic Components

I'm trying to create a model to predict hourly electricity usage. Looking at the data, it appears that there are three different components that I'm going to want to capture in my model. First, there ...
2
votes
0answers
22 views

How to model timeseries temperature data?

I would like to model a timeseries consisting of internal temperature data of a greenhouse, collected at 15 min interval and then use it to make predictions in the future. This is how my data looks ...
0
votes
0answers
16 views

Comparing completely different predictive models

I have a list of different models to predict a chemical value called COD that is based on other parameters (lets call them P1, P2, P3,...Pn). For example: model 1: COD = 1.08P1 + 87*P2 model 2: ...
0
votes
0answers
8 views

Situation based predictive and explanatory models

So the question I am about to ask can be very subjective however I will try my best to ask it in a way that will generalise based on different situations or datasets. I am not comparing predictive ...
0
votes
0answers
17 views

least squares approximation to predict weather

I have daily temperature and rainfall data of fifteen years. I do not know much about stats. So here is my question. How do i use least squares approximation to predict temperature of at least three ...
1
vote
1answer
43 views

Algorithm for weather prediction

I am trying to build a weather prediction app using c#. I am not a stats major and i am trying to understand which simple algorithm can be used to predict temperature and rain fall. I have gathered ...
5
votes
3answers
115 views

Overview of predictive modelling, machine learning, etc.

My text Intuitive Biostatistics is a nonmathematical explanation of conventional statistics. Chapter 3 explains the basic mindset of statistics as analyzing a sample to make inferences from a ...
2
votes
1answer
34 views

How to choose probability to predict success in logistic regression?

I'm working through a logistic regression example from the lab on logistic regression in Intro to Statistical Learning. When they try to test how accurate their model is they do, ...
0
votes
0answers
9 views

Which model to use to handle multiple levels(10-12) in many categorical independent variables and a continuous dependent variable?

I have a continuous dependent variable. 5 categorical independent variable with 7-12 levels in each. Converting into dummy variables and using regression doesn't sound good as there will be so many ...
2
votes
1answer
88 views

Getting hints from target variable: Will it ruin predictive power completely?

I have a large set of predictors and a target variable which is extremely difficult to model. After a couple of failed trials (glm, DT, RF, NN) I got the impression that it is almost random noise. ...
1
vote
1answer
42 views

How to prevent overfitting when encoding categorical variables

Currently, I am working on a binary classification project that include both numeric and categorical variables as predictors. I recently read an article about encoding variable with weight of evidence ...
2
votes
1answer
20 views

How to design a classifier when new data has missing values

I have trained a classifier for medical data, which works ok. Now I have to build a final product to give to the MDs (a sort of program where you give a new patient's record as input and the ...
3
votes
2answers
242 views

Regression with skewed data

Trying to calculate visit counts from demographics and service. The data is very skewed. Histograms: qq plots (left is log): ...
1
vote
0answers
14 views

How to predict the units sold (or a distribution) from a sample of serial numbers?

A thought topic was presented to my probability class years ago (and I've long since forgot) how to determine the final unit sold (or a distribution) from a sample of serial numbers. Example: You go ...
0
votes
0answers
7 views

When making out-of-sample predictions, is there a reason to keep observed values for the predicted variable?

the dependent variable of my investment model has a large number of missing values. Fortunately a variable with a strong relation to the investment value is available for all observations, so that I ...
0
votes
0answers
12 views

Combining results from tests after re-shuffling data

I am fine-tuning a neural network for a binary multilabel classification problem. Basically I am trying to predict 64 binary labels for each input. However, my dataset is somewhat small for the task ...
0
votes
0answers
25 views

Ran a logistic regression and fit my data to the predicted model but the predicted model is not overlapping the observed data

EXPERIMENT: Heaviness judgments of a variable weight to a standard weight (2000g). Independent variable/covariate: Weight (9 levels: 1920g, 1940g, 1960g, ... , 2060g, 2080g) Dependent variable: ...
0
votes
1answer
55 views

How to workout statistical significance

My stats has become very rusty so I’m trying to use this study as a way to bring myself back up to speed. Situation Analysing website visitors to find out at which groups of people prefer to buy a ...
2
votes
1answer
33 views

Curse of dimensionality: why is it a problem that most points are near the edge?

ESL, p. 42 says: Hence most data points are closer to the boundary of the sample space than to any other data point. The reason that this presents a problem is that prediction is much more ...