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
3
votes
3answers
250 views

Deep Learning based time series forecasting

According to the paper "Statistical and Machine Learning forecasting methods: Concerns and ways forward", it looks like the recent DNN-based approach has weaker predictive power in extrapolation, i.e. ...
1
vote
2answers
75 views
+100

Predicting walking routes using PyTorch

I'm working on a project that uses sensors to monitor a persons location. These devices simply record the current GPS coordinates and ping them back to a server (the coordinates will then be converted ...
0
votes
1answer
13 views

How are missing data handled in Time series estimation?

I am looking for most popular/theoretically sound methods for handling missing data in time series model (particularly ARMA class) estimation. Also what method is used in R (in arima and in forecast ...
0
votes
1answer
17 views

Multi-step ahead forecasting with LSTM neural network

I would like to forecast the heat load of a district heating network given its past values, the temperature and the 3-day ahead forecast of the temperature with an LSTM RNN. The data is hourly and I ...
0
votes
0answers
101 views

Irregular seasonality defined as white noise?

I've got data of which I think it has a seasonality. My data has a peak in july/august and one in december. I have only data of 2014 and 2015, but in both of the cases this is happening. (See my graph)...
8
votes
1answer
5k views

Forecasting irregular time series (with R)

There are several methods to make forecasts of equidistant time series (e.g. Holt-Winters, ARIMA, ...). However I am currently working on the following irregular spaced data set, which has a varying ...
4
votes
0answers
556 views

Irregular Seasonality in time series

I understand seasonality of a time series normally means a cyclic component with constant frequency. For example, the frequency is 24 for daily cyclic trend of hourly data. One of the basic models ...
0
votes
1answer
17 views

Forecasting Process with Limited Historical Data and High Variance

I have a general inquiry regarding a project I am working on. I cannot reveal too much, but I would like to gauge the community here and hopefully be pointed towards the right direction in terms of ...
1
vote
1answer
127 views

Is it possible to model continuous time series with exogenous regressors?

I've got an irregularly spaced time series with regressors. I've found the R packages cts and ctsem for continuous time series, but they don't allow for exogenous variables. Is it possible to have ...
0
votes
1answer
208 views

Dealing with missing data in Time Series or non-constant time intervals for forecasting in R (ARIMA, Holt Winters, Theta)

I have a time series of sensor data from a machine. This machine is sometimes moved and thus there are big chunks of missing data, here is a plot of the data points: My goal is to try to start ...
1
vote
1answer
48 views

Forecasting recurring orders for an online subscription business using Facebook Prophet and R

I am analyzing data from a subscription model, in which a customer must pay a recurring price at a regular interval (30 days) for access to the product. EDIT -> Direct link to daily data: https://...
0
votes
0answers
15 views

Code for best ARIMA Model using RMSE in Out Of Sample [on hold]

Reading the book "Introductory Time Series with R" from Paul Cowperwait I found a code that would select the best arima model (similar to using the auto.arima function). So I am interested in ...
1
vote
0answers
28 views

How do you evaluate bias and/or quality of time-series forecasts

I am working on a financial model that will forecast the revenue a company generates over a fiscal quarter, and I am not sure of the best way to rigorously evaluate the bias in the model. Every day ...
1
vote
3answers
280 views

XGboost for Time series - using lag of target variables

I'm trying to make a time series forecast using XGBoost. I have already added many time related variables - day_of_week, month, week_of_month, holiday. I want to add lagged values of target variable ...
0
votes
1answer
37 views

Modeling non-linear (short) time series and cross-validate them

beginner data scientist here. Time series analysis is a completly new area for me, so please correct me if i write something that makes no sense. I have many multivariante short time series, between ...
1
vote
1answer
162 views

Log-Transform/Pre-Processing Time Series before Similarity Matching

I have ~1500 time series data representing store sales (US$). All time series are of the same size with 52 weeks of data with no NA values. For a subset of 18 specific time series, I want to find the ...
0
votes
0answers
18 views

how to calculate safety stock from output of ARIMA model?

I have built an arima model using monthly sales as input suppose the output from ARIMA model is : How do we calculate safety stock for different lead times lead times (in days)?? ...
1
vote
1answer
29 views

multivariate time series: selecting a predictive model

I have a time series dataset that looks like this ...
1
vote
1answer
55 views

Chances a forecasting model exceeds/deceeds a specified threshold [closed]

I am interested in determining the confidence of a forecasting model with applications to quantitative finance. I have the following multivariate data $X$: \begin{align} X(t) \sim F_{X}(t) \end{...
1
vote
0answers
23 views

How to get quantiles/probabilities of time series forecasts?

my problem is as follows : I am creating demand forecasts for some goods with different methods (ARIMA, ETS,..) The issue is that I would like to forecast the probabilities of those forecasts since ...
2
votes
0answers
16 views

Seasonal ARIMA- non stationarity after differencing and seasonal differencing

I am working with a seasonal time series, which is initially stationary. After many attempts, the best model that fits the data is an ARIMA(0,1,4)(0,1,1)[12]. However, checking for the stationarity of ...
1
vote
1answer
250 views

How does facebook prophet handle missing data?

The Prophet paper (forecasting at scale by SJ Taylor - 2017) says the following on missing data: Unlike ARIMA models, the measurements do not need to be regularly spaced, and we do not need to ...
0
votes
0answers
13 views

Forecasting time point not value

I have a simple question. when we want to forecast a time series, we always focus on the value of series in future. But could we forecast time point of spesific value? For example I would like to ...
0
votes
1answer
40 views

How can I quantify the impact of the lag between two events that each occurred twice?

There's a question about the impact of holding sales close to one another. Last year, Sale A was held four weeks before Sale B. This year, Sale A was held three weeks before Sale B. How would I go ...
4
votes
3answers
265 views

Applying an interaction term to all the IVs

I have a linear model with 6 IVs and would like to analyze the effect of an interaction term applied to all the IVs. To illustrate, let's say we're predicting the Win/Loose ratio of NBA basketball ...
0
votes
0answers
293 views

What is the best model for forecasting if you have very less data points? [duplicate]

I recently got to work on a problem of forecasting five years of data. But I only had five data point from previous years i.e. yearly data (frequency = 1). The data is heteroskedastic. For example, [...
0
votes
0answers
17 views

Predictive model fusion informed via observed values

Suppose I have some quantity I want to forecast, like the traffic at a particular intersection or the sales volume at a particular store. I have three sources of data to make use of: Broad-scale ...
0
votes
0answers
12 views

How to Encode Dummy Variables into a Neural Network

I am currently creating a neural network(LSTM) for electrical demand forecasting and I want to include dummy variables to tell the model to treat weekdays differently from weekends, treat working ...
0
votes
0answers
10 views

dimensionality reduction using SVD for forecasting with machine learning

I'm using a LSTM model to forecast time series data. My dataset has far too many variables and I would like to perform dimensionality reduction. My LSTM model works on a rolling window of 500. I ...
2
votes
1answer
36 views

How to predict weekly or monthly sales from daily time series model?

I've been given daily data and I've trained a SARIMAX time series model in Python so that I can predict daily data if given daily input. However, I need to forecast on a monthly or weekly level, ...
0
votes
0answers
6 views

Single Variate Fixed Period Lagged Regression

I found a relationship that seems strong, but I'm not finding corroboration of it in research papers, so, am I missing something obvious? I have data (for simplicity of explanation) ...
1
vote
1answer
24 views

Training ARIMA based on overlapping hourly weather forecasts

I am working with hourly water level data and I plan to forecast each day the next two weeks (on an hourly base meaning 14*24 = 336 forecasts each day). My regressor is an hourly weather forecast. ...
1
vote
1answer
21 views

Median-based Versus Average-based forecast? Which is better?

Hi Cross Validated Community, When generating forecasts (e.g., product-customer time series data), should we choose an average-based forecast or median-based forecast? I recently read a very nice ...
0
votes
1answer
227 views

How to align two seasonal time series

I am trying to decompose a time series using Holt Winters method and use it for forecast. I am trying to do this for weekly data of last 25-26 months. The challenge is that the dates of the seasonal ...
1
vote
1answer
223 views

Models that train on Mean Absolute Error or similar?

I'm trying to do time series prediction and I'm interested in training on MAE or other custom loss functions. For my problem I'd prefer having errors of {0, 10, 0, 10, 0, 10} as supposed to {5, 5, 5, ...
4
votes
3answers
95 views

Is it a valid claim, that by differencing a time series, it loses its memory, and as a result its predictive power?

Marcos Lopez de Prado seems to be a well known and renowned machine learning expert in the field of finance. I am very far from his level, as have not yet finished my PhD in economics, and only have ...
2
votes
1answer
44 views

How to interpret straight line as forecasting

I would like to make some short term forecasting using an AR(I)MA model. having the following daily time series, which is for the raw data: It seems to be like a white noise, based on the acf and ...
0
votes
2answers
60 views

Forecast sales and then ungroupto find individual sales

I am trying to solve a problem for a brewery: A brewery has 50 beer types in total out of which only 8 to 10 beers are available on tap for a single day i.e only 8 to 10 beers will be sold on any ...
4
votes
1answer
47 views

Do out-of-sample fitting methods solve the problem of over-fitting?

Suppose we have a regression model, and we want to fit this to training data, and then make predictions on test data. There is a well-known danger that out-of-sample predictions will be poor, due to "...
0
votes
0answers
20 views

Convert prediction for differenced Time Series ARIMA(1,0,1) and ARIMA (1,1,1)

I am working on a Time Series model, and the series appeared to be non-stationary (presence of trend). I tried 2 ways: 1) put original data into ARIMA(1,1,1) 2) manually difference first order ...
1
vote
0answers
22 views

How to Choose Error Distribution For Time-Series Model

I am modelling a set of time-series, and understand various models (ARIMA, AR, GARCH) allow for the inclusion of non-Gaussian error distributions. I am aware that, after fitting a time-series model, ...
0
votes
1answer
18 views

How to Recursively Predict a Time Series Using Neural Networks

I am currently using neural networks to forecast an electrical demand time series. I am trying to create a forecast for the following day given previous observations at half hourly intervals. My ...
1
vote
2answers
58 views

Understanding MASE Value

I've looked through many of the other posts concerning the Mean Absolute Scaled Error (MASE) forecast metric and haven't been able to sort out my problem just yet. I'm working with some weather ...
3
votes
1answer
168 views

Predictor for averaged Brownian motion

The best forecast (predictor) for a Brownian motion at time $t+h$ is the present value at time $t$ since it's a martingale. The same holds for random walks with independent steps and without drift. ...
0
votes
1answer
40 views

Forecasting high-frequency electricity data with multiple seasonalities

for a school project I need to forecast high-frequency data using different methods of my choice. Data: I have hourly data on day-ahead electricity prices and a few other variables (hourly power ...
1
vote
2answers
50 views

Trend & Seasonality Determination in Time Series without looking at Graph [duplicate]

Most of the articles I have read describe determining Trends and Seasonal (TS) effects through rolling your eyes on Graphs. Graph is a nice visual representation, but I am looking a way in either ...
1
vote
1answer
240 views

Root causes of error in a forecast consisting of two multiplicative factors

I have a warehouse that packages units and ships them. Any number of units can go into the same package, including only 1 unit. I have a forecast for number of units and units per package (UPP). From ...
0
votes
1answer
229 views

Demand bottom-up forecasting and substitution effect

If retailer has many products the is likely to be a substitution effect within product groups (clusters). Hence, there is a notion of the "unit of demand" that is supposed to gather products based ...
0
votes
1answer
262 views

Which algorithm for forecasting a binary time series? [duplicate]

I would like to write the code to forecast the status. The status 0 means non-active, 1 means active. I would like to predict the future month (e.g 2016/6/1), the status should be "0" or "1". What ...
0
votes
0answers
34 views

Big Mart Sales Prediction Problem

I hope that some of you are familiar with Big Mart sales prediction data that was provided by Analytics Vidhya as a contest. The problem statement of on the website is as follows: The data ...