Time series are data observed over time (either in continuous time or at discrete time periods).

learn more… | top users | synonyms

1
vote
1answer
21 views

Interpreting BATS model information using forecast package in R

I have a forecast object in R. When I look at the summary I can see 'Model Information: BATS(1, {1,1}, -, -)' What do these numbers in the parentheses stand for?
1
vote
0answers
32 views

A (simple) example where LSTM works but a regular Neural Network (NN) fails?

In the spirit of the answer from maple on this thread: Using RNN (LSTM) for predicting the timeseries vectors (Theano) I created some simple sine wave data to fit with a LSTM. It worked well! ...
0
votes
0answers
6 views

What statistical test should be used to compare two independent samples over time (pre- and post-test) for binomial variables?

A typical question is: Do you have a paid job? 1 Yes 2 No I have a treatment group and a comparison group (not matched--in fact they're quite different from each other) and I'd like to statistically ...
0
votes
0answers
23 views

Random Forest for time series

I am learning about random forest approaches and was wondering if it is suitable for multiple time series all used to predict the same response variable. For example, three different methods to ...
1
vote
1answer
28 views

Find cointegrating vectors and loadings from a trivariate VAR(1) equation

I have the following process: $\begin{bmatrix} X_t \\ Y_t \\ Z_t \\ \end{bmatrix} = \begin{bmatrix} 1 \\ 0 \\ 1 \\ \end{bmatrix} + \begin{bmatrix} 0.5 & 0.5 & 0 \\ 0 & 1 & 0 \\ 0 &...
0
votes
0answers
12 views

Question about trend stationarity

I am using Matlab in order to test stationarity of macroeconomic time series such as GDP, PCI or population. My first question is the following: I start by testing the stationarity of the data: I ...
-1
votes
1answer
35 views

Writing a function n R [closed]

Hello, How Can i write this function in R? and any simulation codes about SETAR model including codes, links, books or any guide, would be appreciated.? Thanks in advance
1
vote
1answer
20 views

Interpreting Seasonality Component Exponential Smoothing Models

I am building an exponential smoothing model that has seasonality in it, I would like to analyze the data with the seasonal factor removed so I can tell if a performance one month was due to seasonal ...
0
votes
2answers
56 views

Detecting Ramping Events for Wind Power Time Series in R

I'm working with a dataset that spits out a power generation reading from a wind turbine every 5 minutes. Because of the variability in wind speeds the plotted time series are very jumpy, with lots of ...
0
votes
0answers
23 views

How to extend the separated trend line to predict future time series values in R?

We have data from border router devices that depict the bits per second from the past 4 years. There are some missing values or values too low when backup devices were in use. Every day has a BPS ...
0
votes
0answers
33 views

Using RNN (LSTM) for predicting one feature value of a time series

I have been reading several papers, articles and blog posts about RNNs (LSTM specifically) and how we can use them to do time series prediction. In almost all examples and codes I have found, the ...
1
vote
0answers
27 views

How do you define long and short run in an ARDL model?

I am writing up my regression analysis of an ARDL model which includes the long run equation and the short run dynamics. My reader however, would like to know what I mean by long run and short run. ...
0
votes
0answers
4 views

Bounding a Correlation Series

I was wondering what techniques exist for bounding a time series to take values between -1.00 and 1.00. I am asking because I am working with multivariate GARCH models (such as the Dynamic ...
1
vote
2answers
56 views

Arima and lm not giving same coefficients in R

I'm fitting an arima(1,0,0) model using the forecast package in R on the usconsumption ...
0
votes
0answers
30 views

Interpretation of the ARIMA coefficients in a time series

I am trying to understand the coefficients retrieved when placed the command auto.arima to my monthly time series of the annual change in House prices. When doing ...
0
votes
0answers
30 views

What statistical analysis for identifying sections of

I am a stats novice, and dont quite know the method to use in my problem. So I have a set of independent variables, and I want to find what sort of prediction strength exists with some dependants. ...
0
votes
1answer
60 views

Time series prediction using ARIMA vs LSTM

The problem that I am dealing with is predicting time series values. I am looking at one time series at a time and based on for example 15% of the input data, I would like to predict its future values....
1
vote
2answers
72 views

Time series cross validation by reversing the series

I am trying to forecast revenue of a company, using neural networks. The response is a time series of monthly revenues from 11/2008 to 05/2016, and there are about 45 predictors (including lagged ...
1
vote
2answers
75 views

Forecasting: residuals from seasonal decomposition appears to be highly auto-corelated, why?

I am using a publicly available data Kaggle: Rossmann Store Sales and trying to forecast sales. I am using Python. My timeseries is stationary, confirmed via the Dickey-Fuller test. However, I wanted ...
0
votes
0answers
13 views

Confidence interval of fitted value and forecasted value form tbats model in forecast package in R

I have half hourly multi-seasonal(daily, weekly, quarterly, yearly) time series data and I divided them into training part and testing part. ...
1
vote
1answer
22 views

AR1 term significant in ARMA(2,1) but not in ARMA(1,1) or ARMA(1,0)

I'm building an GARCH model using function garchFit from "fGarch" package in R. When specifying the mean equation I have some difficulties understanding what's ...
0
votes
1answer
35 views

Comparing volatility among time series-GARCH Models

I have 10 time series on which I want to compare their volatility by using ARMA-GARCH models. I have estimated the ARMA-GARCH models by using eviews. According to my supervisor now, I should ...
0
votes
1answer
40 views

Applying different time series models (ARIMA, HOLT-WINTER) on the basis of MAPE

I have a time series object calc_visit_ts. I want to apply the best fit time series model based on the MAPE value for each model. The issue I face is that the MAPE ...
0
votes
0answers
22 views

Log transform in time series

After taking the $\log(1+x)$ transformation on a time series, I am guessing which features should I use as predictors: $\text{mean}(\log(1+x))$ vs $\log(1+\text{mean}(x))$ $\text{std}(\log(1+x))$ vs ...
1
vote
0answers
21 views

Plotting Just the Seasonal Component of ETS Model - R [migrated]

This is probably a simple question but my R skills are still in the learning stages. I am trying to get just a plot of the seasonal component in an ETS time series model and I also would like the x ...
-1
votes
0answers
30 views

Time Series Novelty Detection

I am performing very frequent analysis on many datasets of one variable, 15 minute increment, time series data. I want to check if the most recent observation of each dataset is anomalous. The various ...
1
vote
0answers
52 views

Time series prediction in R over more than 180,000 past data points takes forever?

We have data values pertaining to BPS (bits per second) traffic on a networking device. We have data from for a particular month (October) from the past 4 years. The data points are available in a 1 ...
0
votes
0answers
28 views

Rescaling the Implse-Response Function of a Structural Vectorautoregressive Model

I'm thinking about the Impulse-Response Function (IRF) of an Structural Vectorautoregression (SVAR). According to the Enders Book we can write the IRFs with help of the Vector MA-Representation as: \...
0
votes
1answer
38 views

Forecasting multivariate time series data stream

I have a multivariate time series data stream. I am looking for a method that can forecast the next value of one of the variables as the data comes in. (It would be a major advantage if there's an R ...
3
votes
1answer
33 views

Can we use Box-Ljung as a stationarity test for time series?

It's all in the title, I know that we usually use Box-Ljung to test the randomness in a time series (independence of residuals), but I found this post about how to tell if a time series is stationary ...
1
vote
0answers
91 views

Identifying lagged effects / Distributed Lag Model

I would like to create a linear distributed lag model in order to do some forecast and also being able to interpret the results. Unfortunately I'm a bit confused with the process I should follow....
3
votes
0answers
73 views

Splitting data for train/test for time series

A week ago or so I was at a conference. Long story short, I ran into a friend who is quite good at machine learning so I asked them a question about why I might be getting what I think is poor fit on ...
1
vote
1answer
35 views

Solving for a difference equation for a time series variable

I am trying to solve for the values of a variable $u_t$. $u_t$ is defined as: $(1-L-L^2)u_t = \epsilon_t$ where $L$ is the lag operator and $t=1,...,n$. $\epsilon_t$ is a random variable (normal[0,1]...
0
votes
0answers
53 views

Time series prediction in R where data is available over past 4 years in 1 minute intervals

We have data from 'october' in the past 4 years and we want to predict what data for this year is going to look like in October. The data looks like this: 1 2 3 4 5 6 ... Every october has 31 days, ...
0
votes
1answer
24 views

Are spectral decompositions of time-series useful for modeling/forecasting, or are they more of a tool for analysis?

This is a bit of a theoretical question. I'm also new to time-series analysis, and trying to learn fast. Sorry if some of my terminology is off. You can loosely categorize methods to analyze and ...
0
votes
0answers
3 views

Reading time series data into R when you don't have time labels? [migrated]

We have a data set that is in a very strange format. It mentions the start date and the end date and time and then goes into data values. See the format sample below: ...
0
votes
0answers
10 views

small number of treated and control units, with time series

I'd like to examine treatment effects in a before-after randomized experiment where the number of treatment and control units is small, but where I do have a long time series of data for each unit. ...
0
votes
0answers
25 views

Is there a nonparametric test available in R to analyze serial dependencies for short time series?

I am developing a small web application with R and the Shiny-package to conduct easy analysis of single-case time series data. For this purpose, I need a nonparametric test that allows to analyze ...
1
vote
0answers
31 views

Test difference between two short time series

I have data on leukocyte aggregation (a percentage) in two different conditions (A= control, B= Verum). Blood was sampled 8 times throughout the day in 2h intervals. I'm fairly new to statistics and ...
0
votes
0answers
29 views

A time-series analysis problem

I am trying to run an analysis where I am looking at the time it takes for a building lot to go from having submitted its first construction job application to having earned its first construction ...
0
votes
1answer
46 views

Does combining forecast models produce good prediction intervals?

In this blog post, @rob-hyndman says: If you only want point forecasts, that is the best approach available in the forecast package. It is also better than any of the commercial software (at least ...
2
votes
0answers
18 views

Comparing time series of small N group of individuals but reasonably lengthy time series

I have been given some data about pre and post release behavior of a small group of rehabilitated animals in Africa, with some measures of success being how much food the animals eat in the wild and ...
2
votes
0answers
26 views

Is autocorrelation not worth addressing with small N?

Consider a simple regression context in which there is a small set of response values, $Y$, and corresponding dates, $X$. (For simplicity, we can assume the dates are equally spaced.) We would like ...
5
votes
2answers
116 views

Using HAC standard errors although there might be no autocorrelation

I'm running a couple of regressions and, as I wanted to be on the safe side, decided to use HAC (heteroskedasticity & autocorrelation consistent) standard errors throughout. There might be a few ...
1
vote
1answer
65 views

How to interpret TBATS model results and model diagnostics

I have got a half hourly demand data, which is a multiseaonal time series. I used tbats in forecast package in R, and got ...
3
votes
1answer
37 views

Short time series forecast [duplicate]

We have a series of monthly data for 20 months (it is not possible to obtain more data). These are the number of medical consultations encoded in a public hospital and we only have monthly data for ...
0
votes
0answers
7 views

Make auto.arima in R search wider model space [migrated]

I'm using the auto.arima function in R, however I believe it's not searching a wide enough state space, I don't want to set stepwise to false because I dont have the resources for that, however is ...
0
votes
1answer
24 views

Strategy to help classifiers cope with ambiguous examples

I have a machine learning problem where sometimes the training data will correctly have two or more similar/same training examples with different class labels. As an over-simplified example, let us ...
1
vote
1answer
41 views

Using a function of intervention time as a covariate in arimax

I am trying to estimate the impact of an intervention on a time series. The problem is that the intervention effect looks curve-linear (i.e. increasing, plateauing, and then decreasing), and I am ...