Tagged Questions

1
vote
2answers
84 views

Overfitting when using corrected AIC for model selection

I am using the corrected AIC to select the lag order in a simple AR(p) model. I chose the the AICc since my sample is fairly small (n=135). The AICc minimal model is the AR(15). To me it seems like an ...
0
votes
2answers
62 views

State space models for time series forecasting

I am new to time series forecasting and have been slowly working my way through the different approaches available. I've so far mainly been using ets and arima models available in the R forecast ...
0
votes
1answer
34 views

What is the proper name for a backward forecast?

Suppose in time series you have the data in a recent period and you would like to use that data to extrapolate backward to get estimates of the time series back in time. What do you call that? ? ...
1
vote
1answer
43 views

when to aggregate when time series forecasting

I have a some historical sales data for various product SKUs, including category information ("department" "category", "subcategory"). I want to use this to generate sales curve (a baseline forecast ...
6
votes
1answer
118 views

Can you compare AIC values as long as the models are based on the same dataset?

I am doing some forecasting in R using Rob Hyndman's forecast package. The paper belonging to the package can be found here. In the paper, after explaining the automatic forecasting algorithms, the ...
1
vote
2answers
118 views

Forecast total for a year given monthly time series

I have a monthly time series (for 2009-2012 non-stationary, with seasonality). I can use ARIMA (or ETS) to obtain point and interval forecasts for each month of 2013, but I am interested in ...
2
votes
1answer
73 views

How to forecast hourly data in R

I have hourly login data for a web site. Certain hours of the day for example between 09:00 and 12:00, there are heavy traffic on the site. I would like to forecast the hourly data for about one year. ...
2
votes
1answer
43 views

What do error bounds in forecasting represent?

What do error bounds actually mean in forecasting timeseries? For example, when I get a forecast I get the 85% and 95% high and low error bounds. I can also set my own error bounds to be calculated ...
0
votes
0answers
30 views

Implementing an ETS and ARIMA forecast

I've been using the R Forecast package which I have used to create fitted ETS and Arima models. I can easily predict ahead within R using the Forecast package but need to be able to do prediction ...
4
votes
1answer
112 views

Time Series Forecasting with Daily Data: ARIMA with regressor

I'm using a daily time series of sales data that contains about 2 years of daily data points. Based on some of the online-tutorials / examples I tried to identify the seasonality in the data. It seems ...
1
vote
1answer
24 views

How to accurately track the 75% quantile in a non-stationary timeseries?

I have a non-stationary timeseries with a mean (ยต) and standard deviation (SD) which both vary across time. The distribution of the timeseries is skewed, so the left and right sides of the ...
3
votes
1answer
73 views

Approaches to Forecasting with Daily Timeseries

I have just started to learn about forecasting. I thought it would be easy to create forecast models for a daily time series but have encountered a number of difficulties. Firstly most examples and ...
0
votes
2answers
75 views

Need a clear and simple auto-regressive model example

This may be hard to find, but I'd like to read a well-explained auto-regressive model example that: uses minimal math extends the discussion beyond building a model into using that model to forecast ...
2
votes
0answers
19 views

Predict binary occupancy vector from history of vectors

I have a set of binary vectors where each vector represents one day of occupancy in a house and consists of 48 elements (each element for 30 minutes of the day). Each element can be 1 meaning that ...
0
votes
0answers
27 views

How to specify newxreg in prediction model of ARIMA? [migrated]

I have fit the model below to my time series data. The xreg consists of a time vector that goes from 1 through 1000 and of 12 indicator variables (1 or 0) that ...
0
votes
0answers
20 views

forecasting export - methods

I have Product X Export data (time series data: year - amount) approx for last 10 years for my country and also Product Export data for Enterpise Y. I am writing thesis. This would not be the main ...
1
vote
0answers
46 views

Forecasting of highly correlated time series

In time series forecasting using various models like AR,MA,ARMA, etc, we usually focus on the modeling of the data in the change of time. But when we have 2 time series that Pearson correlation ...
0
votes
1answer
47 views

Can I use xreg with stl decomposition to handle moving holiday?

I am trying to decompose and forecast a weekly time series which is believed to be affected by moving holidays (e.g. Chinese New Year, which happens in different weeks of a year). I would like create ...
2
votes
1answer
97 views

ARIMA forecast with seasonality and trend, strange result

as I am stepping into forecasting with ARIMA models, I am trying to understand how I can improve a forecast based on ARIMA fit with seasonality and drift. My data is the following time series ( over ...
0
votes
1answer
48 views

One step ahead forecast with SEASONAL data collected sequentially

In this post it was asked how to do one step ahead forecasts using Arima form the forecast package. Now I'm using an example with hourly seasonal data and would like to do something similar but the ...
2
votes
0answers
79 views

Interpreting time series decomposition using TBATS from R forecast package

I would like to decompose the following time series data into seasonal, trend, and residual componenets. The data is an hourly Cooling Energy Profile from a commercial building: ...
1
vote
0answers
76 views

Simulate forecast sample paths from tbats model

Using the excellent forecast package by Rob Hyndman, I came across the need to not only have prediction intervals, but to simulate a number of future paths, given past observations of a time series ...
0
votes
1answer
54 views

MA on a non-stationary time series

All, I have some data I would like to do some simple forecasting on. Its is non-stationary, looking at the time plot & from ADF & KPSS tests. After differencing I now have a stationary ...
2
votes
1answer
109 views

One step ahead forecast with new data collected sequentially

Hi all I'm trying to do one step ahead forecast. Lets say I have 1000 data and fit an ARIMA model with it and then I do a forecast for one period ahead. When I get more data I would like to forecast ...
0
votes
2answers
89 views

Time Series Analysis and Forecasting

I am looking at ways to forecast monthly time series data over a larger geographic region. I have time series weather data (e.g., temperature, precipitation) from multiple stations, and the stations ...
1
vote
2answers
63 views

Time-series autocorrelations all positive

I've got 36 months of timeseries data, and eyeballing it, it has a linear trend upward. I wanted to do a little more than just eyeball it though. So I put together a correlogram of autocorrelation ...
2
votes
1answer
56 views

What happens to an (AR)MA model when doing out-of-sample forecasting?

What happens to the error terms in an (AR)MA model when doing out-of-sample forecasting? As I understand it, when doing an in-sample fit, the estimate is simply the residual of the ground truth data. ...
-1
votes
1answer
47 views

Data set for forecasting [closed]

I am looking for a data set which can be used for ARIMA or any forecasting models. The data should be such that, over a period of time the range of inputs change i.e. the band of input data changes ...
1
vote
1answer
42 views

Empirically validating a forecast distribution

I have a family of models that give me a forecast distribution for the next observation in a time series. So given observations $O_1, \dots, O_T$, I can calibrate the model and get a distribution for ...
1
vote
2answers
50 views

Puzzled by derivation of time series prediction based on its log

From Introductory Time Series with R: If the random variation is modelled by a multiplicative factor and the variable is positive, an additive decomposition model for $\log(x_t)$ [where $x_t$ is ...
2
votes
1answer
89 views

Forecasting model inputs that are both auto-correlated and are calibrated over time?

How does one account for model inputs that are both a) auto-correlated and b) calibrated over time? I'm interested in forecasting the outcomes of sporting events. Let's say that each team has a score ...
0
votes
1answer
33 views

how to impute missing monthly climate data using AMELIA package?

I have been trying to impute future monthly climate data based on sets of downscaled data using the AMELIA package. I have monthly precipitation data from 1960-2099 for different geographic regions ...
2
votes
2answers
264 views

Using exponential smoothing to forecast irregularly spaced data in R

I'd like to use exponential exponential to forecast the following data. The data is dailly based. Because of some policy reasons, every 29th,30th and 31rt of each month, the data will drop to just ...
3
votes
1answer
328 views

Cannibalization of product sales

I am trying to determine the rate of cannibalization of product sales for A with product B. I am using ~ 2 years of daily sales data for product A and then ~8 months of data for product B. That is, ...
1
vote
3answers
220 views

Correlation between two sets of time series data

I have an assignment to forecast snow depth at a certain location at a specific date. My approach to forecast this is to basically take previous years snowfall data and compare it to this years data ...
2
votes
2answers
123 views

Where can I find time series data to assess accuracy of forecast?

I created an algorithm for forecasting time series (mix of ML methods). Now I need some data so I can compare my results with others and assess accuracy. Unfortunately, I can't find anything like the ...
1
vote
1answer
87 views

Using regression splines for values outside of the calibration range

in my question on a load forecast model using temperature data as covariates I was advised to use regression splines. This really seems to be a/the solution. Now I face the following problem: if I ...
1
vote
1answer
82 views

Time series prediction when data is not i.i.d

I have time series data $y_t$ with covariates $x_{1,t}, x_{2,t}, ...$. The covariates represent budgets for different programs. I can create an ARIMAX model that fits the data very well so far. In ...
4
votes
2answers
355 views

Time series forecast in R with yearly frequency

I have a time series with daily observations over the course of multiple years (interest in topic "superbowl" over time). The seasonality in the data is yearly as well and it is very spiky (almost ...
3
votes
2answers
159 views

Ensemble time series model

I need to automate time-series forecasting, and I don't know in advance the features of those series (seasonality, trend, noise, etc). My aim is not to get the best possible model for each series, ...
0
votes
0answers
94 views

samples from forecasts of VAR time series model in R

I'm trying to do a power analysis for a future experiment with time series financial data. We'll be splitting the data by random (actually, stratified) geographies, so we have a control and ...
0
votes
1answer
77 views

How to regress a time series of proportions?

Every month, an organization surveys some of its customers (the total number of customers is also known). The sampled customers answer a survey with a dozen or so questions; sometimes, customers ...
0
votes
1answer
440 views

Using the R forecast package with missing values and/or irregular time series

I am impressed by the R forecast package, as well as e.g. the zoo package for irregular time series and interpolation of missing ...
0
votes
0answers
54 views

Bayesian Forecast with Minimal data

The following very simple forecast has been very helpful to me in applying basic methods.. and I have read that Bayesian methods may be superior for small data forecasting but I have not seen any ...
4
votes
1answer
103 views

Research Methodology on Fareless Bus System

I am working on a project for a Masters Project. The town I am looking at Switched to a Fareless system in Feb 1, 2011. I want to look and see if this increased ridership by a substantial amount. I ...
3
votes
1answer
120 views

How to forecast number of donations?

I am trying to forecast/predict the number of donations expected to be received at multiple locations across the country. I have and want to use the information I have. This information covers the ...
3
votes
1answer
101 views

Time series regression - ML estimation

I have a linear regression model with some correlated errors: $Y_t=\beta_0+\beta_1X_1+\beta_2X_2+\epsilon_t$, where $\epsilon_t$ is a AR(1) i.e. $\epsilon_t=\phi\epsilon_{t-1}+\nu_t$ with $\nu_t$ as ...
2
votes
1answer
394 views

Cross-validation with neural networks yielding worse results than a standard neural network

Summary: when using a 10-fold cross-validation procedure where each training set is used to generate N bootstrap samples for processing with NNs. How do I provide my NN with correct sequence and ...
4
votes
2answers
178 views

Time series forecasting lookback windows — sliding or growing?

Are there any good reasons to prefer a sliding model training window to a growing window in online time series forecasting (or vice versa)? I'm particularly referring to financial time series. I ...
2
votes
1answer
107 views

Conditional model using function tslm in R package forecast

I would like to use tslm with data that has intraday seasonality and a different pattern on business days and on non-business days. If data.ts is my time series then I would like to use something like ...

1 2 3 4