Questions tagged [forecasting]

Forecasting is the process predicting future values based on historic and current data, typically for time-series datasets.

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

Which machine learning technique can be used for predictive log analysis

I have log data with 100k records. And These parameters. It looks like this. message types can be helpful for anomaly type detection. Out of total 15 message 5 ...
user avatar
0 votes
0 answers
8 views

Forecasting the price of an electricity stock each hour for 3 days with the help of weather data?

I'm new to forecasting and I want to know where can I find information that help me in choosing an appropriate model for this need. What is your opinion on this given that the model needs to predict ...
user avatar
0 votes
1 answer
11 views

Is there a way to forecast a time series multiple linear regression using externally made dummy variables?

This question concerns question 4h of this textbook exercise. It asks to make future predictions based on a chosen TSLM model which involves an endogenously (if i'm using this right) made dummy ...
user avatar
  • 21
1 vote
0 answers
13 views

Discrete wavelet transform - DWT (beginner)

I recently stumbled upon this article : https://www.bportugal.pt/sites/default/files/anexos/papers/wp201612_0.pdf In the paper they use DWT and I am having trouble understanding how to construct them. ...
user avatar
0 votes
1 answer
16 views

Removing seasonality in time series forecasting

In time series forecasting we are removing the "seasonal" component to fit models better and have better forecasting. But why? if I should give an extreme example: if I have a sin wave, I ...
user avatar
0 votes
0 answers
7 views

Confidence/prediction interval vs lower and upper bounds in ARIMA_PLUS big query

How is CI intervals in ML.forecast and lower-upper bounds in ML.Detect_Anomalies different for Arima_plus model in BigqueryML ?
user avatar
0 votes
0 answers
7 views

Can preprocessing in time-series data (e.g. deseasonaliation or detrending) helps create better forecasting model?

I am reading a paper that mentions the following. ...
user avatar
  • 131
0 votes
0 answers
7 views

Multi-level timeseries forecasting? How to do it?

So, I just finished a 48 hr datathon, and I did terribly, to be honest. It was my first datathon. We were given a list of datasets: 5 months of taxi demand data (January to May) Weather dataset Zone ...
user avatar
1 vote
1 answer
19 views

How to incorporate predictor variable without future information into a model?

I will use an extremely simplified example to ilustrate the question, but I think the answer shsould hold for more generalised cases. Let's say I want to create a time series regression model (the ...
user avatar
  • 11
1 vote
0 answers
21 views

How to Visualize attention weights in a Attention based Encoder-Decoder network in Time series forecasting

Below is one example Attention-based Encoder-decoder network for multivariate time series forecasting task. I want to visualize the attention weights. ...
user avatar
0 votes
1 answer
38 views

Predict prices with tensorflow

This is a sample of my dataset: ...
user avatar
0 votes
1 answer
18 views

Dealing with inputs of different sizes in time series forecasting

I'm dealing with a task where I need to forecast the n-ith value of a target variable in a multivariate time series. But in this case we have two variables: -var1: Is my target variable that ...
user avatar
  • 127
0 votes
0 answers
14 views

Multivariate time series forecasting and LSTM: When should I separate time series in different inputs

Let us suppose that I have a multivariate time series with two variables that vary together in time: var1 and var 2. And let us suppose that I want to forecast the n-ith value of var 2, by considering ...
user avatar
  • 127
0 votes
0 answers
18 views

Is there a word for complex knowledge that is not easily integrated into models?

I am currently looking at time series forecasting use cases in my company, and find myself wondering how complex world events, like trade wars, Corona shutdowns and local politics can be integrated ...
user avatar
  • 111
0 votes
0 answers
46 views

Predicting Opportunity Win/Loss using Machine Learning

I have a dataset as below. These are closed opportunities where we have the outcome (won/lost). I want to predict whether the opportunity would be won/lost based on these features and also the time ...
user avatar
  • 235
0 votes
0 answers
33 views

best trial always found at first optuna trial

I am using optuna as part of the pytorch forecasting library. I executed the following code: ...
user avatar
  • 101
1 vote
0 answers
10 views

Adjust loss for values prediction with difference ranges

how do you deal with loss where the goal is to predict sales on different ranges. Let's say i have 2 products ( 1 with a sale of 100 and the other one with a sale of 10000 per day ). Is there a way to ...
user avatar
1 vote
0 answers
26 views

RStudio: Stretched time series cross validation and using exponential smoothing models

I've been have trouble with a question for an assignment. I'm trying to apply an ETS model onto a stretch_tsibble in a time series cross validation. Running this either takes a long time to do or ...
user avatar
  • 21
7 votes
1 answer
462 views

Are Machine Learning Weather Prediction models better than classic weather forecast?

We all know that, there are weather prediction models and case studies. But I don't understand the reason, why people trust them rather than weather forecast on TV. I mean, what is advantages of ...
user avatar
  • 73
0 votes
0 answers
10 views

Making my dataset stationary increases SampleEntropy score

When i make my dataset stationary by the taking difference method, my SampleEntropy score is increasing. It means my data is being "less forecastable". But my results are deffinetly better ...
user avatar
  • 23
1 vote
1 answer
49 views

Explain MAAPE (Mean Arctangent Absolute Percentage Error) in simple terms (intermittent demand forecasting)

n order to measure the accuracy of highly intermitted demand time series, I recently discovered a new accuracy measure, that overcomes the problem of zero values and values close to zero, when ...
user avatar
0 votes
0 answers
7 views

Interpretation of VAR model: about impulse function and lag of p

For example, I have three time series, Y,X1,X2. After using time series cross validation and utilizing BIC/AIC to determine the best p as the lag of the VAR model, in which I got p = 1 to estimate the ...
user avatar
0 votes
0 answers
29 views

Using differencing multiple times for making dataset stationary

When I do differencing to my dataset I am having a lot of zeroes and it causes my prediction to go wrong. But when I use it multiple times, my dataset is having minus values but still, at least my ...
user avatar
2 votes
0 answers
57 views

MinMaxScaler makes my prediction flat

I am trying to do univariate forecasting. But when i try to use MinMax Scaler my predictions are being flat (tried to use different activation functions) but when i use Standart Scaler my predictions ...
user avatar
  • 23
1 vote
0 answers
18 views

How to forecast time series with negative trend in test set and big uncertainty? (uncertainty due to Covid and Ukraine crisis)

Recently I started to create a machine learning model for a European customer for around 800 product time series. The goal is to produce a monthly forecast for the 6 months ahead. Since this customer ...
user avatar
0 votes
2 answers
46 views

Time series test data dilema

I’m trying to build a model to predict the amount of sales of a product for the next few days This question is about whether or not I should use the tail of the serie as the test set and train models ...
user avatar
0 votes
1 answer
14 views

Testing for leading indicators in time series forecasting

Say I want to forecast a time series x. I believe that the values of time series a, b, and, c could be used to predict future values of x but I don't know the magnitude or reliability of their effect ...
user avatar
1 vote
0 answers
52 views

Multivariate Time series forecasting (Sales prediction) using ML techniques

I am looking for pointers to get started on a use case where I want to predict the Sales for t+1, t+2, t+3 using multiple features which may impact the Sales. I do not have data for these features for ...
user avatar
  • 235
0 votes
0 answers
10 views

What is the difference between 2-dimensional input and 2 1-dimensional inputs for LSTM model?

Say I want to build an LSTM model to predict temperature. I have historical data of temperature and precipitation. I could go 2 ways as far as how I construct neural network model (assuming I use ...
user avatar
0 votes
0 answers
10 views

How to use forecasting models

I have built a forecasting model that takes number of users in the previous 30 days to predict for the next day. The model is looking good so far but I actually want to predict for the next 60 days. ...
user avatar
0 votes
0 answers
3 views

timeseries analysis: number and tipology of services requested at t+1 time

I have temporal data regarding the number of customers who requested a specific service in a given period (month and year). Below is a small excerpt from the dataset: Month-year: month and year when ...
user avatar
0 votes
0 answers
12 views

time series forecast, dependent variable is binary

I have time series data, where the dependent variable is binary - either 0 or 1. The 0 value means failure, it's rare, and I want to see if I can get close to estimating the times when it will happen ...
user avatar
0 votes
1 answer
29 views

Where can i learn time series data forcasting/analysis?

I would like to learn time series data analysis and forecasting. I am knowledgeable in machine learning and have a good knowledge of deep learning (including RNN's, LSTM). I came to know that time ...
user avatar
1 vote
2 answers
31 views

Distribute forecasted weekly totals to constituent days of the week

I'm going through Google's Unnoficial Data Science blog, specifically Our quest for robust time series forecasting at scale. Their approach to forecasting includes making weekly total forecasts, and ...
user avatar
  • 249
0 votes
0 answers
11 views

Method for predicting future state, based on time spent in previous states

So what I'm looking for is the best approach to predict a future state. Say we have three states: A, B, C. I want to predict if in the next time-interval (f.e. a day or a week) the state will become C....
user avatar
  • 146
1 vote
1 answer
30 views

How to explain global time series models to a customer?

For one of my customers I need to explain the concept of global models in simple words. Searching for simple introductions to the concept failed so far. All I can find are scientific studies, mostly ...
user avatar
0 votes
0 answers
9 views

What is the best methodology for forecasting multiple time series?

I have $n$ ($\approx$ 800) time series, each of them represents a client, and the information that they carry is daily monetary expenses. I need to develop a methodology to forecast (more than one ...
user avatar
  • 101
0 votes
1 answer
60 views

When R2 score and MSE are not correlated

I'm training some forecasting models and then, to check performance I see several metrics. It's surprising for me when they are no related, for example: Let's suppose I'd have two models, A and B. --&...
user avatar
  • 101
0 votes
1 answer
57 views

Running model.fit multiple times for an LSTM?

I have time-series histogram data from many separate machine runs (see this post for detail). I am working to train an LSTM in order to predict the final histogram in a machine run based on the past ...
user avatar
0 votes
0 answers
24 views

Seasonal random walk formula, Forecasting Principles and Practice

Seasonal naïve method: A similar method is useful for highly seasonal data. In this case, we set each forecast to be equal to the last observed value from the same season of the year (e.g., the same ...
user avatar
0 votes
0 answers
22 views

Multiple time series forecasting method

Your help will be very useful in the below exercise as I have issues in my try to identify the correct ML approach to solve it. My target here is to predict the value of the of 23:00 hour for the last ...
user avatar
1 vote
0 answers
12 views

Fail to decompose and make stationary time series

I am looking for some suggestions for my time series. I am dealing with the column "Temperature (C)" from this dataset. I am trying to make it stationary in order to do some forecasting on ...
user avatar
  • 169
0 votes
0 answers
9 views

How to properly measure forecast errors when predicting correlation coefficient?

My task is to accurately predict correlation coefficient value. I have some candidate models, and want to select the best one (with minimal forecast errors on validation dataset). I don't feel good ...
user avatar
1 vote
0 answers
33 views

Price Predition for Irregular spaced historic data of non independent Prices

I am a little unsure how to proceed. I am not an expert but on a decent intermediate level when it comes to regular Timeseries. Now i am faced with a problem that first seemed related, but is an ...
user avatar
1 vote
0 answers
38 views

How does Kalman filtering work backwards?

For time series missing value imputation I am using the Kalman filter/smoothing approach, given in the imputeTS package. As Kalman filter is iterative and needs a view data points to make its estimate ...
user avatar
1 vote
0 answers
30 views

Estimating sales for new products in e-commerce

I am trying to find an ML solution for the following problem. Objective: Given a list of products of Stores B, C, D … estimate the Order Value (1 year timeframe, let’s say) that each product would ...
user avatar
  • 11
2 votes
0 answers
22 views

Why does a linear regression in time series forecasting does not provide a line in predictions?

I'm reading the TensorFlow Time Series forecasting Tutorial 1 trying to perform my own time series prediction. However, specifically on single-shot models section for multiple time steps, the Linear ...
user avatar
0 votes
0 answers
43 views

When should I use neural networks?

I am struggling with this exercise. The objective is "to build a recommendation system that predicts the next video" viewed by a user, given the data provided. So, the dataset consists in ...
user avatar
0 votes
0 answers
15 views

How to train ARIMA model on multiple similar time series?

I am having 'business potential' values of 4000 cities (having generic names to ensure anonymity) for 72 months. The data for an individual city is just 72 months so I clustered the entire dataset ...
user avatar
0 votes
0 answers
42 views

Time Series Analysis for Categorical Data Output

Suppose I am having dataset which consist of date as one column and fruits as second column which is categorical data having set of 4 different fruits in that column and my output column has 0's and 1'...
user avatar
  • 36

1
2 3 4 5 6