Tagged Questions

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

learn more… | top users | synonyms

1
vote
0answers
14 views

Selecting an appropriate VAR model

I would like to receive critical comments on an idea explained below. Suppose I have variables $x_1$ through $x_K$, and this is a time series setting. My aim is to forecast variable $x_1$. I know ...
1
vote
0answers
32 views

Stationarity ⇒ homoscedasticity? [duplicate]

If my data is stationary, can I also write that it is homoscedastic? Does stationarity imply homoscedasticity of the data?
0
votes
0answers
8 views

Basic questions about stochastic gradient descent / Robbins and Monro algorithm

I have a LOT of time series observations and I would like to estimate a simple AR(1) model $$ y_t =c+ \phi y_{t-1}+ \varepsilon_t \qquad \varepsilon_t \sim \text{N}(0, \sigma^{2}) $$ with parameters ...
-2
votes
0answers
17 views

Matlab: How to calculate Expectaion for the random variable [on hold]

I have a time series model y(t)= h^T y(t-1) + n(t) where n(t) is a white Gaussian noise that excites and drives the process. y ...
7
votes
4answers
179 views

Analyzing up/down patterns in short time-series data

I have not worked very frequently with time-series data, so am looking for some pointers as to how best proceed with this particular question. Let's say I have the following data - graphed below: ...
2
votes
1answer
52 views

How to implement model in R?

i would like your help to implement this model in R or more explicity where yt = monthly mean values μi = mean value in month i, i = 1 . . . 12 . I1;t = Indicator series for month i of the ...
1
vote
0answers
17 views

How to compute the out-of-sample log-likelihood function?

I am doing some empirical work based on realized GARCH model, whose log likelihood function is given as $$ l(r,x;\theta) = -n\log(2\pi) -\sum_{t=1}^n \left[\log(h_t) + \frac{r_t^2}{h_t} + ...
1
vote
1answer
33 views

proving the asymptotic distribution of the mean

Let ${X_t} = \mu + \sum\limits_{j = - \infty }^{ + \infty } {{\psi _j}{\varepsilon _{t - j}}}$ with $\varepsilon$ is a white noise iid with variance $\sigma^2$ , $\sum\limits_{j = - \infty }^{ + ...
0
votes
0answers
20 views

what is the best method of smoothing time series for product share data?

I am having a share data for products presribed over period of time.The share is calculated like for 3 products each one 1/3 share and like that,where products may vary and hence their share. What is ...
0
votes
0answers
9 views

Obtain the graph of the autocorrelation function in ARIMA models [migrated]

I am implementing an ARIMA model in Python for forecasting U.S. GDP. I am interested in obtaining the graph for the autocorrelation function. I obtained the values for ACF but I can not see the ...
0
votes
0answers
18 views

Finding genuine arrears and default arrears from rent payment patterns

I am currently working on some housing data - in particular analyzing the tenants' rent payment information and I am stuck on progressing with the following: I have to classify tenants based on their ...
1
vote
0answers
8 views

Coarse-resolution subsampling of time-series data

Suppose I have time series data with a very fine resolution, e.g. 100 datapoints per second. I want to report this data to some service that can only take data at 1 point per second. I need to do ...
0
votes
0answers
10 views

Linear time series model and z transform

I am few questions from the following papers: ...
0
votes
0answers
20 views

Techniques for comparing two windows of data in a time series

I'm working on a small independent project in R, trying to make my own (very crude) forecasting method. The general idea of the component that is giving me trouble is trying to compare two windows of ...
1
vote
0answers
10 views

Training/ Test Data with Time Series Model — Forecast with Training Model, or with Model based on Full Data?

Okay, I have a couple books on time series forecasting, but perhaps I need to read a couple more. Here's my question. You want to be able to validate a forecasting model. So you split the data into ...
0
votes
1answer
26 views

R times series — correct use of forecast() and accuracy() in forecast package

Cross-posting this from Stack Overflow, because it's a bit of a stats/ technology cross-over. I'm relatively new to R and the forecast package I believe authored by Rob Hyndman. I'm having trouble ...
1
vote
1answer
51 views

Analysis of residuals

For my master thesis I have implemented following forecasting models: naive (just to check) decomposition method exponential smoothing (single/double/holt-winters) SARIMA Now I need to do the ...
1
vote
0answers
22 views

Using A Time Series To “Scale” Another

I know the "average theoretical cost per impression" for Jan 13 - Dec 13. I have other monthly time series for "total # of impressions", "total # of clicks" and "total number of conversions" for Jan ...
1
vote
0answers
16 views

Prediction on univariate time series. How?

Suppose you have more than one univariate time series. Each of them shows how a variable y changes over the day (e.g. one reading per minute) and refers to one of N different geographical areas (for ...
0
votes
0answers
8 views

Given two time series where one is dependant on another how do we find the spectrum or acvf in order to then find the spectrum.?

$X(t) = aX(t-1) + Z(t)$ and $Y(t) = bY(t-1) - X(t)$ how do you find the spectrum if $a=b <1$ and if $a\neq b \neq 0$ and $a,b <1$?
1
vote
0answers
22 views

Hold out sample vs. cross validation for time series, and how to perform in R

I think out-of-sample validation testing for accuracy is essential in initially judging what time-series forecasts to use. In any case, I've been doing some reading on the two most common methods, ...
0
votes
0answers
9 views

Optimal length of observation window

I have a sample of N stocks, with time-series of daily returns. For each stock, I would like to compute the sample univariate variance of returns, using a rolling window. These variance will be used ...
0
votes
0answers
23 views

autocorrelation in evaluating time series forecasts

I'm having some trouble wrapping my head around whether using Holt-Winters ETS or an ARIMA model for forecasting sales figures (which are highly seasonal). I'm been using R and the Forecast package ...
0
votes
0answers
13 views

VAR model for price forecasting in multiple time-series context. How to get “real figures” as forecasts?

Sorry for the rather long introduction, but since I was (legitimately) critizised for not explaining my cause and questions enough, I will do so now. I would like to conduct a (price)-forecast based ...
0
votes
0answers
11 views

handling trend in predictor and response variable

I am trying to create a linear regression model containing two predictors and 1 response variable. My response variable has a short term pattern, i.e. surge during weekdays and slump during weekends ...
-1
votes
1answer
27 views

What does the Argument “type” in VAR() - function do?

Right now I am working with vector autoregressive models in order to make 3 months forecasts for a commodity good (sawlogs) y. I have several time-series of "follow-up-products" of sawlogs that should ...
0
votes
0answers
13 views

Predicting the near-future values using an unevenly sampled time-series data

Summary Need help with predicting the near-future values using an unevenly sampled time-series data. Data is collected as events, and is converted to time series. I have tried out a few approached ...
0
votes
0answers
15 views

creating an indexed dummy variable as a predictor in OLS

I am performing on OLS with two predictors and a response variable. The data is a time series of 450 days approximately. There is an irregular pattern in my response variable - it sometimes ...
0
votes
1answer
16 views

How do I deal with missing data for a repeated measure collected over time?

I have a real-life data set, with only one "measured" variable (i.e., patient waiting time). This single variable is collected weekly, in different clinics, for different providers, across time. I am ...
0
votes
1answer
47 views

Transfer function in forecasting models - interpretation

I am occupied with ARIMA modelling augmented with exogenous variables for promotional modelling purposes and i have hard time explaining it to business users. In some cases software packages end up ...
1
vote
1answer
33 views

Time series modelling

Here is my problem: I basically have 20 or so variables (I have 1000 of these values over an increasing time axis). I want to calculate the weights of these input variables. I am going to try Linear ...
0
votes
0answers
20 views

time series definition for time between failure

I recently started on forecasting time-between-failure for failure of different components in a truck. I saw a few papers which used ARIMA to do the forecasting for number of failures at specific time ...
3
votes
1answer
55 views

Is there a way to plot an average a set of time series that minimises 'temporal smearing'?

Let's say I have n time series datasets, each of which displays a very similar pattern—perhaps an essentially flat line with a pronounced bump somewhere in the middle—and I want to display them all ...
2
votes
2answers
60 views

Why monthly stock returns instead of daily returns in multiple regressions?

This is probably a naïve question. Why do many multiple regression analyses of the Fama-French 3 or 4 factor model of fund returns use monthly return data instead of daily return data? I would have ...
0
votes
0answers
24 views

forcing certain parameters to be skipped during optim in R

I have a code which tests each possible order of ARIMA and selects the best model by choosing the one with the absolute minimum sum of lags from the PACF graph. The code then proceeds to add weight to ...
0
votes
0answers
23 views

Time Series Analysis Approach

I've got a dataset for the time series analysis. I used HoltWinters method in R, but the results were not satisfactory. The HoltWinters method was used with gamma set to False. The fitted plot of ...
1
vote
1answer
57 views

Interpretation of mean absolute scaled error (MASE)

Mean absolute scaled error (MASE) is a measure of forecast accuracy proposed by Koehler & Hyndman (2006). $$MASE=\frac{MAE}{MAE_{in-sample, \, naive}}$$ where $MAE$ is the mean absolute error ...
0
votes
2answers
38 views

Definitions of coefficients from Arima {forecast}

I'm trying to explain in detail step by step what my code does and I am stuck at explaining what the coefficients are in an Arima model and where they are from/what relevance they have. Could someone ...
0
votes
1answer
20 views

Retrain Time Series Models

I'm new to TS modeling, but have some experience in classic classification modeling. In classification I can train one model and use it for some time while some indices are stable (e.g. PSI). ...
1
vote
1answer
29 views

Distinguishing diffusion from white noise

I have a time series that looks like this: This comes from an experiment, and I know the following: Originally, for $t < s$ the time series is $x_t = vt + e_i$, where $v$ for this particular ...
1
vote
1answer
80 views
+100

Multiple testing in correlation analysis over time periods

I have one variable measured once per time interval (say, once per year), and another variable measured periodically (say, once per day). The periodic measurements are autocorrelated. I am interest ...
0
votes
0answers
19 views

Generalized likelihood ratio test

Does anyone use the generalized likelihood ratio test for detecting a sudden change in time series forecasting (ARIMA Model)? A paper by Bonne Zhu uses this technique for anomaly detection, but I ...
0
votes
1answer
30 views

Multilinear Model with fixed intercept

I would like to fit the following model Y (t) = m (t) + b * t + g * C (t) + N (t) with m (t) to be the long term mean monthly values (remove seasonal component), b ...
0
votes
0answers
26 views

How to estimate trend from only 3 data points

I have annual data for the rate of event $X$ in a population from 2000 to 2013. I've fitted a poisson model to establish an observation period for a trend analysis from 2000 to 2010. I've then used ...
0
votes
1answer
25 views

How to transform non-Gaussian multivariate time series

I wish to apply a VAR-like kind of model to a multivariate time series dataset. The model assumes that $X_t | X_{t-1} \sim \mathcal{N}(\Gamma X_{t-1},\Omega)$ for $X_t \in \mathbb{R}^p$. I want to ...
1
vote
1answer
24 views

comparison of forecasting models for daily data (frequecy=365)

I have 852 days of daily attendance data and need to use the first 800 days data to predict the next 52 days and match it with my actual values. How do i decide which is the best model to use for ...
0
votes
1answer
19 views

Formatting Time-series data for Cross-correlation

In an experiment we measured 100 response times for each subject. The data has the following format: ...
0
votes
1answer
25 views

combining subjective probability estimates and statistical estimates for forecasting

At the end of the year forecasters usually struggle year to predict landing estimate for the financial year due to variety of reasons including volatility, unreliable demand projections, inventory ...
0
votes
0answers
20 views

Hypothesis testing on groups of time series data

I have some experimental data from two groups, where each group contains data from $n$ subjects. The data are in the form of a time series for each subject, but are not all the same length. To be ...
1
vote
1answer
23 views

Avoiding spurious regression with cross-sectional data

I have been reading everything I can get my hands on about spurious regression but can't seem to definitively find out what is best in regressing cross-sectional data including non-stationary ...