Tagged Questions

Forecasting involves estimating the value or distribution of a random variable which has not yet been observed.

learn more… | top users | synonyms (1)

0
votes
0answers
5 views

Event level driven response modeling

I am investigating operational and maintenance data for a fielded system. There is a year worth of data. The operational data has been reduced to fault indications, which are triggered when ...
-1
votes
0answers
13 views

Cran Package from Google for Forecasting multivariate time series [on hold]

Recently, i saw the package that can forecasting multivariate time series and evaluate the effects of variables like promotion at main time series. Please help me to find it again
1
vote
0answers
19 views

auto.arima and Arima (forecast package)

I am facing a strange issue with auto.arima. On a dataset named data, I run the following code ...
0
votes
0answers
17 views

Time series forecasting with multiple series with constraints

Hello and thanks in advance. I am using ARIMA or VAR models to forecast sales revenue. Suppose I have three different time series in each of three categories (making 9 series in total). The first ...
-2
votes
0answers
17 views

Forecasting in R X Axis [migrated]

Good day How do I change the x axis so that it shows the year and month? At the moment the x axis doesn't look right and comes up with 2014.0, 2014.5 and 2015.0. I want to use the forecast package ...
3
votes
3answers
139 views

How do I handle nonexistent or missing data?

I tried a forecasting method and want to check if my method is correct or not. My study is comparing different kinds of mutual funds. I want to use the GCC index as a benchmark for one of them but ...
0
votes
0answers
17 views

how could i handle with missing data or non existent data? [duplicate]

i tried a forecasting method and i want to check if it is correct or not and why? my study is about evaluating mutual funds for two kind of them it is a comparative study and i wan to use gcc index ...
1
vote
0answers
32 views

Regression model for predicting life expectancy

I have average life expectancy at birth data for an 8 year period and I would like to use that 8 year period to predict the trend for average life expectancy for the next 5 years. I would then like to ...
0
votes
0answers
5 views

error for xreg and newxreg matrix size in an arima model

I am using auto.arima to forecast a daily data and used holiday and weekday dummies as regressors. I am getting an error which does not make sense. When I want to predict a 7 step ahead forecast, it ...
1
vote
0answers
29 views

Why only full ARIMA models in auto.arima?

It seems that the auto.arima() function in the forecast package in R only considers full ...
0
votes
0answers
21 views

Adding predictor variables and/ or systematic judgement to time series forecasts

I have a ways to go with my forecasting general education --- but I'm doing a seasonal time-series forecast for predicting sales order volumes. It's mostly software sales, which does have a ...
1
vote
0answers
25 views

Does the density of daily data impact forecast accuracy?

I know it might be trivial but does the density of daily values impact the forecast accuracy? For example, if a call center receives less than 50 calls for weekdays and less than 10 calls for weekend, ...
-2
votes
0answers
36 views

why cannot we find an arima model for the daily forecast of a dataset with more than 2 years of historical data? [closed]

I faced this problem a few times. I used TBATS to fit a model to the daily forecast of a variable and then wanted to include other regressors to my forecast. So I used auto.arima. all of my databases ...
0
votes
1answer
84 views

Theories how they arrived at this Ebola growth forecast? Not regression

The 21,000 estimate for Oct. was certainly not via quad or power regression. I wonder how they got that number? ...
0
votes
0answers
17 views

time-series forecasting - predicting av. error intervals

I should start with the disclaimer I'm not proficient with R or heavier statistical terminology, sadly! Nevertheless I create sales forecasts (using different methods such as holt-winters), and I'm ...
1
vote
0answers
72 views

In-Sample vs. Out-of-Sample One-Step Ahead Forecasts

I'm fairly new to forecasting but I find all of this quote fascinating and hope to learn something from all of you. I have 500 observations and I'm tasked with the following: "compute recursive ...
0
votes
0answers
47 views

How to implement a simple Bayesian Network for Time Series Data?

I'm a computer science grad student, with not much knowledge in Bayesian statistics, so I'm seeking for guidance for the simplest start. I have 10 variables, like demand, price etc. and I want to ...
0
votes
0answers
30 views

What is the source of nonstationarity in this VAR model?

I am trying to forecast a VAR model, which consists out of 5 variables with a monthly frequency. The problem is that the VAR model produces an unstable forecast and I am not sure what the source of ...
0
votes
0answers
15 views

Auto.arima choose between lots of regressors

I have to forecast data with two seasonality with ARIMA. I find that I have to use a code like this: ...
1
vote
1answer
76 views

ARIMAX with a specified nonlinear model using the arima function in R

I am interested in fitting an ARIMAX model using R. As known, ARIMAX can be understood as a composition of ARIMA models and regression models with exogenous (independent) variables. I have a time ...
0
votes
2answers
67 views

What's the minimum sample size required to do a time series analysis?

I'd like to know the minimum number of monthly data points required to do time series analysis with the seasonality effect in forecasting. I read some articles & they were saying that 50 or 60 ...
4
votes
1answer
183 views

Predicting Y from a regression model for dY

I have some time series data where I'm modelling temperature as a function of various predictors. On physical grounds, I can expect that $$\frac{dT}{dt} \propto T_a - T$$ where $T_a$ is the ambient ...
1
vote
2answers
50 views

Time series with autoregressive error

How can I in R fit a time series, $x_t$, with external regressors, $v_t$, and an autoregressive error? This time series model is given as follows, $x_t = \beta v_t + \epsilon_t$ where $\epsilon_t = ...
0
votes
0answers
13 views

VaR using GARCH

...
0
votes
1answer
33 views

GARCH forecasting

...
0
votes
1answer
15 views

What are approaches to, accuracy and value of forecasting in/for highly volatile environments?

More details in my Quora question here: http://qr.ae/x4s5Z. Please note that this question is not about value, approaches and methods of forecasting in general, but specifically about forecasting ...
2
votes
1answer
49 views

How to build a function with the result of auto.arima in R?

I use: fit = auto.arima(Y, xreg=X) in R to get ARIMA(1,0,0), result as follows: ...
0
votes
0answers
27 views

How much training data is enough for seasonal time series forecast

I am new to times series forecast. If I have data(single variable and timestamp) with double seasonality periods, which are 288 and 1056. And I use tbats in R to build time series data and then ...
0
votes
0answers
19 views

forecast package - SSE output missing

I'm doing some exercises (from Hyndman's Forecasting Principles and Practice text) using the forecast package. I'm specifically using the holt command. The Holt/exponential smoothing is optimized ...
0
votes
0answers
23 views

Have a data set for 3 consecutive days. What are my options?

Let's say I have a set regarding the transportation methods(Eg: car, bus, train) used for three consecutive days b y $n$ number of people. For simplicity let us assume that everyone use only one type ...
1
vote
2answers
155 views

very high frequency time series analysis (seconds) and Forecasting (Python/R)

I have high frequency data (observations separated by seconds), which I'd like to analyse and eventually forecast short-term periods (1/5/10/15/60 min ahead) using ARIMA models. My whole data set is ...
0
votes
0answers
19 views

How to simulate a structural break time series? [migrated]

I want to know how to simulate the following structural break autoregressive time series: $\begin{cases} Y_t = 0.9Y_{t-1}+\epsilon_t & \text{for }1\le t< 50\\ Y_t = ...
0
votes
0answers
10 views

what is the best forecast interval for tbats?

I used tbats to fit a model to a set of weekly data (but only for weekdays) so I have seasonality of 5 for daily and 261 of annual. I wonder if tbats works better by updating the model using observed ...
0
votes
0answers
23 views

Parameter estimation for dynamic regression models with correlated noise ARMA errors

I'm reading the Dynamic Regression Models chapter ( https://www.otexts.org/fpp/9/1 ) in Professor Hyndman's book, and I couldn't understand how to fit the regression model when the error is modeled ...
1
vote
2answers
64 views

Why is the intersect negative and what does my regression show

I am trying to get my regression right. I want to see, if subs increase how much increase in revenue is seen. The dependent variable is Revenue while the independent variable is subscribers. Least ...
0
votes
0answers
15 views

updating a forecasting model including the new observed data with the historical data

I want to have a one week ahead forecast for my data which includes a four years of daily historic data (three years are used as train set and the 4th year is used as the test set). I can use three ...
1
vote
1answer
144 views

What if the trend is changed?

I want to forecast tourist arrivals using time series analysis. I expected to use monthly data from 2000-2013. But due to the civil war, the trend was changed after 2008 as in the following plot. ...
0
votes
1answer
25 views

Very different Neural Network test errors for same architecture

So I'm doing a time series prediction, and assessing the capability of the ANN to predict that time series. I am using Matlab's neural network toolbox functions, and the training parameters are the ...
0
votes
0answers
14 views

How forecasting is made

Can someone, please, provide some guidelines how forecasting is made about e.g. the number of TVs sold for future periods, using mobile services, etc.? What are the techniques?
1
vote
1answer
33 views

Is it reasonable to use a combination of two forecasting models for a dataset?

I used tbats to fit a model for a 3 years of historic data and the values work fine but as I did not include holidays, holiday predictions are really off. I used arima with regressor (holidays at ...
0
votes
0answers
20 views

what is the problem that auto.arima can not provide a model with seasonality for a data set which have strong seasonality?

I have 3 years of daily data which have daily and annual seasonality. I used tbats to fit a model but when I included holiday and weekend regressor, the fitted model did not change with tbats with no ...
0
votes
0answers
94 views

Obtaining the SarimaX equation from the arima coefficients

I have a SarimaX model with three regressor variables: ...
0
votes
0answers
22 views

forecasting for data excluding holiday dummies

I used tbats to fit a model to a set of 3 years of historic data for daily number of shipments moved by a trucking company. my data included double seasonality so I used tbats. However, tbats did not ...
0
votes
1answer
56 views

Is there any way to include regressors in tbats function in r? [closed]

I knew that tbtas could not consider regressor in r but I knew that Dr Hydman group were working to include this feature. I just wanted to know if it is still under work. Thanks
1
vote
1answer
51 views

How can one say if a model is poor based on RMSE value

I have a general question about the value of using RMSE to see if a forecasting model is poor. I used the forecast package in R to find forecasting models for ...
0
votes
0answers
22 views

Need to forecast a small data set. Suggest best method to go about

Hi I have sales data for previous 3 years(6 half years). I need to predict / forecast the sales for next 1-2 years. Tell me which method / model I should use. As always sales dependent on country ...
0
votes
4answers
175 views

Can predictive power be inferred from only in-sample modelling results?

I wonder if one can tell anything about predictive power of a model if model selection and estimation was done using all available data. That is, there was no data left for "out of sample" prediction ...
0
votes
0answers
31 views

prediction for data including weekly and annually seasonality and dummy variables for holidays

I have a three years of daily data for number of orders a trucking company receives everyday. Number of orders are high during weekdays and they have a huge decrease for weekend. I used msts to ...
1
vote
1answer
44 views

prediction using historic data with unusual annual trends

I have 4 years of daily data. there is a decreasing trend for the data for the first 3 years but the trend increase for the 4th year. I wanted to find a fitted model using the first 3 years and then ...
1
vote
2answers
268 views

Wrong predictions for weekend, but good predictions for weekdays

I have a set of 3 years of daily data. I saw weekly and annual seasonality in the data so I used msts time series and tbats ...