Questions tagged [time-series]

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

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

Regression with time series as dependent variable

I want to perform a regression between 3 variables [x1,x2,x3] that have no trend and no seasonality across their time observations and a variable [Y] that has trend and seasonality. For [Y] I've ...
0
votes
0answers
5 views

How should I interpret irregular lag correlation in time series?

I'm not sure how to interpret the autocorrelation pattern I'm seeing in this time series and any help would be very appreciated. Below is the ACF plot for my data. To me this looks like seasonal ...
0
votes
0answers
8 views

ARIMA(0, 1, 0) or ARIMA(0, 0, 0) for Stock log-Returns Forecast

I'm trying to forecast the log-returns of Amazon's stocks using the ARIMA model, so I went through the traditional procedure of examining the autocorrelation plot and the partial autocorrelation plot ...
1
vote
2answers
24 views

ACF vs PACF in ARIMA

Given a time series problem, Should ACF and PACF be done before or after differencing that make the time series stationary? If ACF and PACF has shown different results, should the number of orders ...
0
votes
1answer
21 views

What happens when a time series is multiplied by an iid error

Let's say say I have a standard AR(1) process, apart from the fact that is multiplied by the $ε_i$~$iid (0,1)$. Would this affect the independence of the the AR(1) series? My intuition is no ...
0
votes
1answer
22 views

Why does absolutely-summable weights ensures a linear series itself summable (convergent)? Some questions on def'n of Linear Series

A "linear series" $y_t$ is the linear combination $$y_t - \mu = \sum_{i=-\infty}^{\infty}\psi_iL^i\nu_t = \sum_{i=-\infty}^{\infty}\psi_i\nu_{t-i}=S(L)\nu_t $$ of weighted (by $\psi_i$ weights) lags ...
0
votes
2answers
44 views

how to deal with panel data using linear regression

i have a data set that with several columns about the top 5 managers in 100 firms from 2009 to 2018, some managers still the same through years and some had been changed for some firms in some years ...
0
votes
0answers
16 views

predicting future orders based on past ordering data [on hold]

I'm a student and as my project problem statement is as follows: a company has provided ordering data ( SKU, quantity,date of order, customer name) of past 3 years in form of an excel file. we want ...
0
votes
0answers
11 views

how to find a test for detecting fluctuation in a time series?

I have a time series which consists of, say, 100 values (which comes from a loss function). Normally, the loss goes down over iterations. However, at some point it just starts fluctuating. I would ...
0
votes
0answers
21 views

Dynamic regression, models with coefficients = 0 chosen as top models

I am running auto.arima on part of a time series (training data) using all possible combinations for several external regressors. I then choose the top 5 models according to fit to testing data using ...
3
votes
0answers
35 views

Time series predictions look suspiciously good

I am working on a time series forecasting problem. For this, I am training a recurrent neural network in Keras (mostly following the guidelines from this blog post by Jason Brownlee). My problem ...
0
votes
0answers
10 views

Expected value of ARMA(2,1) process

Let $X_t = \alpha_1X_{t-1}+\alpha_2X_{t-2}+\theta\epsilon_{t-1}+\epsilon_{t}+\phi$ with $\epsilon\sim N(0,\sigma)$ Solve: $E_t[\frac{1}{X_{t+1}}]=?$ Does this even have a solution? Appreciate the ...
2
votes
0answers
49 views

Correlations between two sequences of irrational numbers

If $x$ is an irrational number and $b$ an integer, let's define $g(x,k) = \mbox{Correl}(\{nx\},\{nb^kx\})$. Here $k=1,2,\cdots$ is an integer. The brackets represent the fractional part function. ...
0
votes
0answers
8 views

How to find FFT for a window period of a given large time series data?

If i binned time series data for particular time interval 't' and choose a window period of lets say 5 bins and converted into n rows as train data(5 bins for each row) and a y_value(need to be ...
0
votes
0answers
16 views

Time-series decomposition with multiple seasonality

I'm trying to extract the trend of a daily time-series in R. I've tested the 2 following methods, but I'm not sure which one to pick. Are there any criteria I can apply to choose objectively? Method ...
0
votes
1answer
42 views

If $y_t$ is a time series with autocovariance $\gamma$, does $\gamma$ necessarily have to be absolutely-summable?

If $y_t$ is a time series with autocovariance $\gamma$, does $\gamma$ necessarily have to be absolutely-summable; i.e., ${\sum_{i=\infty}^\infty |\gamma (i)}|<\infty$? If not, what could be the ...
0
votes
0answers
10 views

Updating prediction with valid test data

The following scenario must be extremely common but I couldn't find a best practice for it readily available. Suppose we require a predictive model (m) that is supposed to explain some variable y ...
3
votes
2answers
50 views

anomaly detection in time series training data

I have a dataset which basically is measuring the number of people passing a certain region which is monitored and I basically have these raw counts of people over the last two months at 5 minutes ...
1
vote
1answer
17 views

Unconditional mean of ARMA(2,1)

Why is the unconditional mean of this model equal to 0? Can we tell if we don’t know anything about stationarity? I’m self studying and this got me confused. Thank you in advance.
1
vote
2answers
20 views

How do I forecast quarterly public expenses based on annual budgets and potentially other variables?

I have some time series data from 2008 and forward (see below) on quarterly public expenses and annual public budgets. I would like to forecast the last two quarters of 2018 as precisely as possible, ...
1
vote
0answers
23 views

Temperature timeseries agreement significance test

I hope my question wasn't answered before, but after extensive research I wasn't able to find something similar, so I'm asking for help (this is not my field). I have air temperatures timeseries ...
0
votes
0answers
12 views

How to estimate policy impact using interrupted time series, when outcome variable is a moving average?

Suppose you want to evaluate the impact of a recently implemented policy using interrupted time series because you have no comparison group and the policy was implemented all at once on the ...
3
votes
1answer
33 views

What is the correct number of observations to report for an ARIMA/ARIMAX model?

This might be due to my relative inexperience with time series modelling, but I am confused about the correct number of observations to report for an ARIMA/ARIMAX model. I couldn't find any post that ...
0
votes
0answers
5 views

Significance of hyper parameters in the DHR model in R forecast package

The Dynamic Harmonic Regression model in R requires the input of parameters K, the length of which depends on the number of seasonality in the forecast data. According to https://otexts.com/fpp2/dhr....
1
vote
0answers
14 views

VECM in cointegration analysis: How to chose the form of the deterministic terms from dataplots

Please have a look at the plot of 5 different time series $Y_t=(y_{1,t},y_{2,t},y_{3,t},y_{4,t},y_{5,t})$. I want to use R's urca package to perform cointegration ...
0
votes
0answers
69 views

Neural Network regression on time series

I want to predict the trend values of a time serie [Y] based on the effect of other 10 input variables which can also have interaction. Since the combination of interaction between the inputs is ...
0
votes
0answers
18 views

mgcv::bam convergence problems - scaling issue?

My apologies, this is a long question: the TL;DR version is "my GAMs don't converge, is this a distribution issue and if so can I fix it?" Edit: fixed the distribution information. ...
0
votes
0answers
22 views

Investigating influence of multiple time series

I'm trying to build a model, with which I'll be able to see the impact of changing one time series on another. For example: there are metrics A, B and C, I want to use my model to say ...
0
votes
0answers
20 views

Deep learning model for multiple sparse time series

I have a dataset with multiple (about 50) sparse time series (of 30 days), and the data is sparse : most of the variables are like (1,0,0,0,..,2.5,0,0,...0), contains many zeros, and I want to ...
0
votes
1answer
14 views

Validation set for hyperparameter tuning of ML time series model

I'm developing an ML-based model to forecast the daily sales of a whole month. This model takes as input a set of precomputed time series features: day_of_week, <...
0
votes
1answer
22 views

ARIMA Time Series Simulation - Media Mix Model

I have designed and tested a time series model where I am able to examine the impact of various marketing channels on dependent variables (Such as sales, revenue, website traffic, etc). The model has ...
1
vote
0answers
18 views

High frequency FX data and sample size - possible autocorrelation

I have data for EUR/USD at 1 minute intervals for 2003-2018 (approx 5.86million data points) I have read papers that say that "high frequency data" suffers from "extremely high negative ...
0
votes
0answers
26 views

Generating data between two time series observations [on hold]

I have a time series data-set which is monthly and have got just 48 odd observations.I want to generate daily data between each two observation.,i.e. say 30 daily points between two monthly points. ...
0
votes
0answers
18 views

Estimating variance in a random process

Suppose that $(X_t)_t$ is a random process, and that I am given just one realization of it, through the sample $(x_t)_t$. Suppose I want to estimate the variance $var(X_{t_0})$ for a given $t_0$. What ...
1
vote
1answer
34 views

Suggest a model for segmenting a time series

Hi all, I have a time series that looks like above. I'm interested in segmenting it into the numbering listed. I've tried using a hidden Markov model, but the versions I can find with multiple ...
1
vote
2answers
22 views

Can i use short time series data?

I want to run ols regression for time series data in R, but my data is short that is annual from 2000-2009. There are only 9 variables(2000-2009) and i collected data for inflation and exchange rate ...
1
vote
0answers
12 views

How to back transform the foretasted values from Prophet algorithm?

i have log transformed the actual and fed the data into forecasting model(Prophet) and i have the forecast in scaled to log format. I would like to transform them back to normal values. Should taking ...
0
votes
0answers
19 views

Find plausible peaks in streamed data

i have got a signal of a streamed source which produces values like in the picture. I want to get the "real" peaks (blue circles). But the noisy peaks (red circles) mess up the peak search. The ...
1
vote
1answer
24 views

What is a good measure of the similarity of 6 different time series?

Essentially, I have 6 different data time series that were each generated first using an industry standard methodology (call it method m.A) and then again using my technique (call it method m.B). ...
0
votes
1answer
21 views

Should I use a multivariate analysis or N univariate analyses in this case? [closed]

I have 100 Investment funds (Flexible allocation Morningstar category, same investment area, currency and distribution status: the sample is homogeneous) over a 10 yrs period. I want to estimate a ...
0
votes
0answers
7 views

CUSUM-type tests for change in variance: Difference between Sansó, Aragó and Carrion (2004), Kokoszka and Leipus (2000) and Andreou and Ghysels (2002)

I'm trying to figure out the difference between the tests of Sansó, Aragó and Carrion (2004) and the test of Kokoszka and Leipus (2000) after the adjustment suggestion by Andreou and Ghysels (2002). ...
2
votes
1answer
34 views

What is the virtue of loading absolutely-summability in the definition of causality of ARMA model?

An ARMA series $y_t$ is causal function of $\nu_t$ if there exists constants $\psi_j$ such that $\sum_{j=0}^{\infty} |\psi_j|<\infty$ and $y_t=\sum_{j=0}^{\infty} \psi_j\nu_{t-j}<\infty$ for ...
0
votes
2answers
14 views

Arch models: dependence and squared residuals

I would like a mathematical and intuitive answer to those questions: Why a dynamics like arch in the volatility of a time series implies that the time series is dependent (although not autocorrelated)...
0
votes
0answers
11 views
0
votes
0answers
8 views

Convergence warning in LME4 despite p~=0.05 via ANCOVA? [migrated]

I’m having trouble with convergence warnings using lme4. I'm collecting time-series data. The outcome measure ("TCV") is derived from electromyography. I run repeated measures tests on the same ...
0
votes
1answer
14 views
1
vote
1answer
34 views

How to interpret the the results of augmented Dickey-Fuller test to make conclusions about the order of integration

I am following Pfaff 2011 chapter 3 and 5.1 to find the order of integration of a time series $y_t$ by augmented Dickey-Fuller (ADF). Basically what we do here is testing whether $y$ has a unit root ...
3
votes
2answers
58 views

Time Series Forecasting, why try and predict randomness?

Maybe I have a gap in my understanding, but I've looked everywhere and haven't found a satisfying answer. When we are doing time series forecasting, we need the data to be stationary, if not we apply ...
0
votes
1answer
45 views

Proof for how the drift estimator, for a random walk with drift, is unbiased?

Random walk with drift formula is: (Yt = α + Yt-1 + εt ) How do I go about checking that the drift estimator α-hat is unbiased.. which is proving that E(α-hat) = α? Is this something I would need ...
2
votes
1answer
65 views

GAM factor-smooth interactions and model selection [on hold]

I'm working with a dataset which is a long-term animal abundance survey collected from 11 sites, which have different average temperature(also there are some missing years for some sites). Here, the ...