Podcast #128: We chat with Kent C Dodds about why he loves React and discuss what life was like in the dark days before Git. Listen now.

Questions tagged [forecasting]

Prediction of the future events. It is a special case of [prediction], in the context of [time-series].

Filter by
Sorted by
Tagged with
0
votes
0answers
14 views

Do these forecasts imply the ARIMA model is misspecified?

I have a time series of a stock return over more than 2 years. It's stationary (Augmented Dickey-Fuller test is significant). The plot looks like this: The ACF and PACF look like this: I think these ...
0
votes
0answers
9 views

How to understand a time series reached its saturation point or maximum growth?

I am curious about how to find that a time series model reached its saturation point or maximum growth which doesn't change over time and stable for a while? Saturated can also be considered as not ...
0
votes
1answer
28 views

How can I remove trend of model's forecast when I use ARIMA model?

I have to forecast future energy consumption. I decided to use ARIMA model. But my model's forecast shows the wrong trend. The blue line shows true value. And the orange line shows my model's forecast....
1
vote
0answers
26 views

Good results with Linear Regression

I am experimenting with a data set that I have of revenue of few products. It is on a monthly level and I have 6 years of data with me. As a preprocessing step, I've converted that data set into a ...
0
votes
0answers
12 views

M4 forecasting competition models in R [on hold]

Anyone knows if someone manages to compile the different methods submitted for M4 competition into a R package? The methods were submitted separately and have been compiled here https://github.com/...
3
votes
1answer
87 views

Why is time series forecasting different for each software?

I have 2 different software programs: SPSS, and Statgraphics. I am using them for time series forecasting but Each one gives different arima parameters when using the auto ARIMA model, and The ...
0
votes
1answer
28 views

Automated Fourier Term Selection in Batch Forecasting via Foreach

I'm trying to parallelize my batch forecasting via foreach, and I'm pretty comfortable doing this via auto.arima and ...
0
votes
0answers
15 views

Remove season and trend from multivariate tsibble [migrated]

I am using the stl() function to remove season and trend from a multivariate timeseries. I then want to keep to remainder term from the stl decomposition in order ...
1
vote
0answers
47 views

Can I use ARIMA with hour data for two year prediction? [closed]

I am trying to use ARIMA model for time series forecasting. My data consists of hour by hour energy consumption. I have data for one year. So I have total 24*365 observations for energy consumption. ...
2
votes
1answer
32 views

95% prediction interval for an ARMA(2,2) model

What would the formula for a 95% prediction interval for an ARMA(2,2) model be? The specific model I am using is: an ARIMA(2,0,2) with non-zero mean, with the following parameter estimates: ...
0
votes
0answers
9 views

What is the way to find p, d, q value of non-seasonal, and P,D,Q value? [closed]

How to find the p, d, q value of non seasonal and P, D, Q value of seasonal part from a f and pacf plot in tone series Arima model. Auto arima is not providing good results. I want to use Arima model....
0
votes
0answers
27 views

Problem with forecasting in R [closed]

So basically i am working with the japan ir and ipi data from fred database. I did this work on eviews but i wanted to pass it to R. The thing is that in eviews you have the option to choose between ...
1
vote
0answers
19 views

Fitting and forecasting a multivariate time series model (VAR) in R

I have some quarterly time series data for accumulated total public expenses and the total budget that I want to forecast. I also have subsets of the total public expenses, eg. health expenses and ...
0
votes
0answers
27 views

Forecasting intermittent demand with Dynamic Negative binomial [on hold]

In this paper Forecasting the intermittent demand for slow-moving inventories: A modelling approach (International Journal of Forecasting). The authors propose an autoregressive negative binomial ...
0
votes
1answer
14 views

Machine leaning model to predict whether customer would go out of balance or not based on the historical trasaction data [closed]

I want to prepare ML model that would help me in predicting customer's zero balance event using models like time series forecasting or any other techniques. I am getting customers transnational data ...
0
votes
0answers
12 views

Regression for long tailed time series events

I have a set of values which are a time series and follow a long tailed skewed distribution. I would like to understand what the best method might be to predict the next value in the series. Do the ...
-1
votes
1answer
80 views

How can I improve my ARIMA forecast in R?

I obtained the ARIMA forcast below from the following code: ...
1
vote
0answers
14 views

Suggestions for appropriate time series model , continuous outcome, time varying covariates

I am a dealing with a dataset which is as follows ...
1
vote
0answers
66 views

Can I use linear regression for predicting time series? [closed]

I have the percentage of defaulted clients in porfolio for different periods as target variable and macroeconomic factors as explanatory variables. Can I use simple linear regression? What other steps ...
2
votes
1answer
18 views

Acceptable Standard for MAPE

What is the general acceptable value of MAPE in industry ?. I am getting MAPE of around 24% on live data that has 48 data points in which 42 as train data and 6 as test data. I am trying to do ARIMA ...
1
vote
1answer
27 views

Forecasting method to use with large seasonal swing but otherwise stable data

I am attempting to forecast percentage of churn. However, I am running into issues. The churn is fairly stable except at each year anniversary point. For example, data looks like this for the first ...
2
votes
1answer
18 views

Time Series Forecast with Only 2 Observations

I have a number that I need to forecast its distribution over a period of time (5-7 years). There are currently 2 past observations, but I'd need to forecast the distribution of the rest. What options ...
2
votes
1answer
36 views

How does the R function Arima () calculate drift?

The Arima() function in the R forecast package contains an "include.drift" parameter. Could someone explain how this is calculated and how it is included in point forecasts? According to this post ...
1
vote
1answer
38 views

Does a Vector Autoregression Model truly avoids circular function issues?

Let's say you have a VAR model that estimates GDP and Unemployment among many other variables using a certain number of lags. This VAR model can estimate or regress GDP using Unemployment. And, it ...
1
vote
0answers
10 views

Is there a relation between the characteristic roots and the s.e of the arima coefficients?

in Forecasting: Principles and Practice there's a warning that inverse characteristic roots close to the unit circle may be numerically unstable, and the corresponding model will not be good for ...
0
votes
0answers
21 views

How do I weight multiple forecasts based on Mean Absolute Percentage Error (MAPE)?

I have 3 forecasts that have different Mean Absolute Percentage Error values that were averaged for each forecast over a 6 month period at a monthly cadence against end of month actuals. How would I ...
0
votes
1answer
40 views

Time Series Analysis - ARIMA

I am trying to make sales prediction from time series data. After performing a log transformation to the original data and differencing it by 1, I got a stationary dataset. So I plotted ACF and PACF ...
2
votes
1answer
32 views

Alternatives for Diebold-Mariano test when comparing the best forecast among many against a benchmark

Suppose I encounter a new forecasting method and I wish to evaluate it against a benchmark. I can obtain forecasts from the two methods and compare them to actual realizations and thus obtain the ...
0
votes
0answers
36 views

Deriving the general form of the best linear predictor $\tilde{X}_{n+m}$ of $X_{n+m}$ for AR(1) process in terms of $X_1, …, X_n$

I'm trying to derive the best linear predictor $\tilde{X}_{n+m}$ for $X_{n+m}$ for a causal, zero-mean AR(1) process $Z_t = X_t - \phi X_{t-1}$. My answer needs to be in terms of $X_1, X_2, ..., X_n$. ...
0
votes
0answers
8 views

Incorporate sales forecast uncertainty into product stock availability simulation

I work in a e-commerce and I’m estimating availability of our catalog products. The process (in its simplified form) looks the following. A customer order comes in and we fulfill it if we have stock ...
1
vote
2answers
19 views

Extract standard errors from Arima model applied to groups using sweep

I am following the sweep vignette on Forecasting Time Series Groups in the tidyverse, see here. sweep is the ...
1
vote
0answers
14 views

Covariance of prediction errors

I have an exercise to compute the covariance between the prediction errors, but I'm not sure if it is correct, this is the exercise; I have an AR(1) model, $y_t = \phi y_{t-1} + \epsilon_t$, where $\...
3
votes
2answers
45 views

How to forecast number of event occurences in a certain time period?

I am attempting to predict how many times a certain event will happen in a time period. For instance, predict the number of time the event will occur in the next 5 hours. I have data going back a ...
0
votes
0answers
12 views

Are Random Forests limited for forecasting/projection purposes

Today a coworker came asking me if we could train a model that will predict some land prices and then use said model to predict on hypothetical data. Let's say, I have a ...
1
vote
1answer
27 views
0
votes
1answer
21 views

First difference or seasonal difference in VAR/VECM

I have monthly data on house price, rental price, wage index and interest rates. I want to use VAR to produce impulse response function. Is there any reason why I should use first difference, x(t)/x(...
1
vote
1answer
27 views

Log-Log regression and cost function

I have made a very siple linear regression model having used log-log tranformation for the y and one of the independent variables: log(y)=B0+log(X1)B1+X2B2 where B0 is the intercept and B1,B2 the ...
0
votes
1answer
50 views

Are there any method to multivariate forecasting except for VAR?

I mean 2 or more dependent variables as multivariate.I know we can use VAR or VARMA to forecasting but are there another method except for ANN and VAR?
2
votes
1answer
23 views

How to calculate seasonal naive quantile forecasts?

I was reading a chapter on prediction intervals and saw that the standard deviation is used to calculate "prediction" intervals. Earlier today I was reading some results from GEFCom2017 where ...
3
votes
3answers
67 views

What's the difference between Time Series Regression and Forecasting?

I often see the concepts Time Series Regression and Time Series Forecasting refering to something similar but I don't see clearly what's the difference among these two concepts. By now, the idea I ...
2
votes
1answer
132 views

Inventory forecasting algorithm for wines?

I'm trying to figure out a good inventory forecasting algorithm for wine sales. I have the following characteristics so far: Wine sales are cyclical and not seasonal Have incomplete time series: ...
0
votes
1answer
28 views

MAPE and SMAPE shift invariance (bias)

MAPE (Mean Absolute Percentage Error) and SMAPE (Symmetric Mean Absolute Percentage Error) both are sensitive when the TRUE ...
0
votes
0answers
12 views

How to figure out which features indicate similarity between time series?

I have a data set with the following data: A large number of time series of sales for various products. A large set (~100) of static features that describe each of these products (e.g. product ...
2
votes
2answers
66 views

Comparison of MAE and Mean to illustrate the error magnitude

I have predicted a time series with positive, zero and negative values. As error measurement I used the Mean Absolute Error (MAE). In order to give the reader of my paper a better understanding ...
0
votes
0answers
40 views

Prediction of $X_{n+1}$ with Yule-Walker estimate

Consider a causal AR(1) process $X_t = \phi X_{t−1} + Z_t$ with $(Z_t)$ iid with mean 0 and finite variance. I am reading in a book, that $\phi X_n$ is the best predictor for $X_{n+1}$ because it ...
2
votes
1answer
30 views

Prewhitening with seasonal response and non-seasonal independent variable

I'm working to develop a forecasting model for a quarterly seasonal variable (quarterly estimated individual income tax payments) using several candidates for non-seasonal independent variables (...
0
votes
0answers
15 views

how to estimate hyper-parameters when cross-validating time series forecasting?

I want to evaluate several forecasting methods on the taylor time series using cross-validation. How do I go about selecting the hyper-parameters for the methods? ...
0
votes
0answers
15 views

Is it necessarily incorrect to randomize train-test split for a time series random forest model?

As part of some preliminary research, I'm experimenting with a random forest classification model for predicting whether the S&P 500 will be higher or lower at tomorrow's close versus today's ...
1
vote
0answers
28 views

How does BATS model work?

I am using BATS on a univariate time series model. I have observed strange behaviour. I have data from 2016 to till date (weekly level). If actual are considered from 2016 to 2019 May, I have used ...
0
votes
0answers
9 views

RNN (LSTM) training on multiple time series

Regarding RNN training, We feed network a network -> point by point from the same time series (or image or smth else). When we “switch from one time series to another”, what should be done or how ...