Questions tagged [time-series]

A Time series is a sequence of data points with values measured at successive times (either in continuous time or at discrete time periods). Time series analysis exploits this natural temporal ordering to extract meaning and trends from the underlying data.

0
votes
0answers
8 views

Imputed predictions for missing time-series data nearly stationary (flat line)

I have player over time data that is missing player counts over several years. I'm trying to fill in/predict the missing player count data over different intervals. Data available here: https://1drv....
0
votes
0answers
9 views

Future 100 steps prediction inverse transformation error

I have 608 data points. I want to forecast the next 100 data points. I have already done the basic train-test validation & have finalized a model. I have used 8 dummy variables in my model (namely ...
-2
votes
0answers
12 views

plot this type of data as it contain categorical value in Matplotlib or convert it to numeric

This data i have I have a data set that is of timeseries and i want to plot it using matplotlib but the problem is every row contain a alphabets so help me it to convert the whole data to numeric.
-1
votes
0answers
14 views

Error in res.sum$coefficients[2, 1] : subscript out of bounds

Hi i am new to R and i am trying to do Time-series Analysis where there are 2 attributes Dates and Calls handled its a pretty small dataset of 78 entries for the first time. However the issue ...
0
votes
0answers
10 views

How to give datetime as input features in Ludwig Toolbox for Time series forecasting?

I'm Trying for uni variate time series forecasting with Ludwig toolbox.I need to give my date column of type datetime as input. But I am getting error when i try to pass datetime column as timeseries ...
0
votes
0answers
12 views

Passing the time series in the modwt functions gives an error

I have a data frame consisting of various country stock index values on daily basis. I am converting it into time series and then passing it into modwt function for variance calculations. It's giving ...
0
votes
0answers
12 views

Time series on a small dataset. Also application of ADF is not happening

Hi I am new in R and was tring to convert dataframe into a time series object but after applying groupby on a certain index the datatype gets changed to "tbl_df" "tbl" "data.frame" format. ...
0
votes
0answers
9 views

Is it it possible to accelerate the training of the Bayesian structural time series model (BSTS) by splitting the workload on multiple cpu cores?

I'm trying to speed up the training of BSTS model as described in question title. Is this possible? If yes, how would one implement it? The mcmc iterations are independent, right?
1
vote
0answers
19 views

Adding CustomBusinessHour to DateTimeIndex in vectorized manner

Suppose I have a DateTimeIndex which I wish to add a CustomBusinessHour to. My CustomBusinessHour keeps indexes the same that fall between 0900 and 2000 on the same day, it moves indexes that fall ...
0
votes
0answers
7 views

One dimensional CNNs for time series feature extraction

One dimensional CNNs are used in many tasks for feature extraction of time-series data. In many papers, I have read that one-dimensional CNNs are able to extract the most relevant spatial information ...
0
votes
0answers
17 views

Extract trend using x13 is not smooth

Here is my code: X13=sm.tsa.x13_arima_analysis(TEST,freq=12) There is a kink in the graph approximately from May 2008 to June 2008. Can you suggest any tune method or any other method for ...
1
vote
0answers
11 views

Need to CreateLoop for ARIMA (p,d,q)

I need to find a way to run a loop on ARIMA code across multiple files or within a data frame This is for my thesis project i am working on. However; given the sample size - running the code one by ...
-1
votes
0answers
10 views

How can I create a following polar graph with contours?

Polar graph I want to create visualization like above using python. I have time-series data. I have created below visualization. How can I make it look like the above graph? fig=plt.figure(figsize=(...
0
votes
0answers
21 views

If i forecast some data, that display a server workload, ARIMA is not even close forecasting a simular pattern

I want to forecast some data with the ARIMA. Everything works when I want to forecast some testdata, but the ARIMA model is not even close, and fits straight lines or just some quadratic funktions. I ...
0
votes
0answers
22 views

Detecting seasonality given time series data in Python [on hold]

Is there any way to test time series data for seasonality? The output can just be a True/False. How do I implement this: https://robjhyndman.com/hyndsight/detecting-seasonality/
-2
votes
0answers
18 views

Python code to reverse Yeo-Johnson transformation

I am a beginner in time series topics and I am trying to use prophet library for my study. To clean my data, I used yeo-johnson transformation (instead of Box Cox because of negative values). Now, ...
0
votes
0answers
20 views

interpolate after doing outlier using cubic or spline method

I have a problem where i use pandas.core.series.Series.interpolate with cubic method and it's fill everything except the first and the last index Here is my code : train = data.iloc[:, column_no-...
1
vote
1answer
62 views

Skewed Window Function & Hive Source Partitions?

The data I am reading via Spark is highly skewed Hive Table with the following stats. (MIN, 25TH, MEDIAN, 75TH, MAX) via Spark UI: 1506.0 B / 0 232.4 KB / 27288 247.3 KB / 29025 371.0 KB / ...
0
votes
0answers
24 views

Join two time serie data set [on hold]

I got 2 Time series data sets, dataFast(10ms) and dataChg(onChange) which i would like to join, so that i got a complete view over the time. Until now i got this statment: SELECT f.idDataFast , ...
-1
votes
0answers
12 views

How to choose a training set in the case of a time-series data set?

How to choose the best training set in the case I've a time-series dataset ? I have 30 years data day day, so it's a little ambiguous for me to split the data ! I always look for a justification in ...
-1
votes
0answers
31 views

Why sql is throwing error for DateTime in time Series ? Unsupported input data type in column 'DateTime'

I am trying to integrate my Python time series code in SQL server 2017. Written a procedure for that. The input to the procedure is a table having two columns DateTime and Data. Inside the python ...
1
vote
2answers
17 views

Is there a way to resample timeseries data into x hours and get output in One-Hot encoded format?

Is there a way to resample timeseries data into x hours and get output in One-Hot encoded format? I've tried pandas resample with sum function and count functions, but they don't give output in the ...
1
vote
0answers
25 views

Time series with multiple stores multiple products

I have a 300 stores and 20000 products in my data set and I want to predict next 3 months sales forecast for each outlet each product level. And my data frame looks like this (sample) like this data ...
0
votes
0answers
20 views

LSTM Time Series Classificasion val_acc is fixed or less than 0.5

After reading and googling my concern is that there is something wrong with my data. The data come from two sources, with deference intervals. I joined the two sources and re-sampled the data so it ...
-2
votes
0answers
14 views

Dealing with sudden rise in sales for inventory prediction

I am dealing with time series data of sales with a sudden peak in a very short time and am I trying to capture this. I am using fbprophet for prediction. Any idea of how to deal with this situation? ...
1
vote
0answers
36 views

Implement Causal CNN in Keras for multivariate time-series prediction

This question is a followup to my previous question here: Multi-feature causal CNN - Keras implementation, however, there are numerous things that are unclear to me that I think it warrants a new ...
0
votes
1answer
49 views

R^2 score is not well-defined with less than two samples. Python Sklearn

I am using a Linear Regression classifier to predict some values. I already figured the basic part of the out and now it looks like this: import time as ti import pandas as pd import numpy as np ...
0
votes
0answers
11 views

getting error 'str' object has no attribute 'freqstr' when i run the ARMA model on daily WTI crude oil prices

Following is the model, I am running on the WTI crude oil prices model_arma = ARMA(train['Dollars_Per_Barrel'], order = (4,4)) and it throws the error: 'str' object has no attribute 'freqstr'
-1
votes
0answers
14 views

How to shock the dynamic factor models or factor models

I have a dynamic factor model estimated from a set of multivariate time series. Now I have some difficulty to get a clear relationship between the factors and the responses. Let me describe the ...
0
votes
0answers
19 views

Multivariate time series forecasting lstm

Assume I have 200 samples of time series data. The input has three features and I want to forecast one step ahead in time. So assume my series look like the following: The first three rows: 1 2 3 4 5 ...
1
vote
1answer
35 views

Scaling / parametric weighting function for time series analysis

I have some time series data and I would like to weight my data, so that recent observations are weighted higher than older observations. Therefor I'm looking for a parametric weighting function, ...
-1
votes
3answers
59 views

How can I create a rolling mean for each user based on previous 7 days of activity?

I've been looking at past posts and can't seem to find something that matches my needs. Goal: For each user, I want a mean of their previous 7 days of activities (not counting the current observation)...
-1
votes
0answers
17 views

Is there a model named holts winters no seasonal

The exponential smoothing family of models which i know are holts winters, holts winters damped, holts trend and simple exponential smoothing (ets). The code below shows holts winters, holts winters ...
1
vote
0answers
15 views

How to un-log var predict model and apply to original data [on hold]

After creating var predict model crime_var <- VAR(crime, p=2,type = "const", lag.max = 10, ic="AIC") predict_f <- predict(crime_var, n.ahead = 2) fanchart(predict_f) chart is displaying with ...
0
votes
1answer
20 views

How to use Apache beam to process Historic Time series data?

I have the Apache Beam model to process multiple time series in real time. Deployed on GCP DataFlow, it combines multiple time series into windows, and calculates the aggregate etc. I now need to ...
-3
votes
0answers
19 views

Error in the method of Clustering based on SI(silhouette index)? It is actually do-able of not? [closed]

I am trying to make my own clustering algorithm which is based on SI? It basically does is calculate all the SI and nearest clusters and changes its labels if SI is negative. Also maximizing the ...
0
votes
1answer
28 views

How create Multiple Lines Chart with Python in a single graphic from different files?

I'm currently working with many different time series csv files. One file of them contains the observed temperatures for different hours and the n others files contain the predicted temperature from n ...
0
votes
0answers
29 views

How to fix 'non-numeric argument to mathematical function'

I am totally new in R and doing an assignment for my studies. While coding I found an error stating "non-numeric argument to mathematical function". I Know the header is causing it, as it is the only ...
0
votes
0answers
13 views

Run kubernetes job worker queue on cassandra timeseries

I want to use the kubernetes job on worker queue https://kubernetes.io/docs/tasks/job/coarse-parallel-processing-work-queue/ However, I have a Cassandra time-series queue and not sure if there is ...
0
votes
1answer
49 views

How to efficiently map data between time series in python

I'm trying to create an efficient function for re-sampling time-series data. Assumption: Both sets of time-series data have the same start and end time. (I do this in a separate step.) Resample ...
2
votes
1answer
80 views

Plotting 12 monthly time series data: builtin `graphics::plot()` gives 'cannot plot more than 10 series as “multiple”'

I want to plot 12 different monthly time series data over the year range 1984-2018. I want to get a graph like the following but graphics::plot() always gives me an error - what's the solution? Code: ...
-1
votes
0answers
21 views

time series forecasting using seq2seq

I am currently doing a similar time series forecasting for numbers of calls with seq2seq model. Now I meet a problem which is that when I want to predict like 30 days target's values in the future, ...
1
vote
1answer
26 views

Calculate difference sequentially by groups in pandas

I'm trying to calculate the difference between two columns sequentially as efficiently as possible. My DataFrame looks like this: category sales initial_stock 1 2 20 1 ...
1
vote
0answers
36 views

How to interprete the frequency = 7 with 365 daily data?

I have a dataframe with daily stock data over a 1 year history. I want to decompose my time series, so i set the frequency = 7 (otherwise frequency = 365 span the error time series has no or less than ...
1
vote
1answer
29 views

Modelling time series data with explanatory variable

I have the below data, where y is average price of baskets of some products x is the exchange rate. period y x 1 201501 1530 2.49 2 201502 1450 2.62 3 201503 1637 2.77 4 201504 1404 ...
0
votes
1answer
43 views

sort_values function in pandas dataframe not working properly

I have a dataset of 1281695 rows and 4 columns in which I have 6 years of monthly data from 2013 to 2019. So, it's obvious to have repeated dates in the dataset. I want to arrange data as dates in ...
1
vote
1answer
24 views

Calculate change over time with tidy data in R - do you have to spread and gather?

Quick question about calculating a change over time for tidy data. Do I need to spread the data, mutate the variable and then gather the data again (see below), or is there a quicker way to do this ...
-3
votes
0answers
18 views

Is there a Theta forecasting method available in Pyhton [closed]

Is there any available Python Package for the Theta forecasting model (univariate time-series)? I refer to the paper : Assimakopoulos, V. and Nikolopoulos, K. (2000). The theta model: a ...
0
votes
1answer
37 views

Input shape for LSTM which has one hot encoded data

The sample dataset contains Location point of the user. df.head() user tslot Location_point 0 0 2015-12-04 13:00:00 4356 1 0 2015-12-04 13:15:00 4356 2 0 ...
0
votes
1answer
22 views

Correlation between time series

I have a dataset where a process is described as a time series made of ~2000 points and 1500 dimensions. I would like to quantify how much each dimension is correlated with another time series ...