Questions tagged [probabilistic-forecasts]
The probabilistic-forecasts tag has no usage guidance.
19
questions
0
votes
0
answers
11
views
The dependency in the instances of a random variable [duplicate]
Suppose we have a list L of all root words in English numbered from 1 to n. The data is any English text (let's say a text from a book) where each word is replaced by its root. You are given the data ...
- 262
1
vote
0
answers
41
views
Do "types" of predictions "need" to match assumptions about data?
Let $\mathcal X$ (observation space) and $\mathcal P$ (prediction space) be non-empty sets.
Let $P$ be a variadic function (predictive model) mapping tuples of elements of $\mathcal X$ into $\mathcal ...
- 11
0
votes
0
answers
21
views
Can you do Quantile regression for probabilistic forecast with Gaussian process regression (GPR)?
I am learning probabilistic forecasting and there are three way to do it, quantile regression, prediction interval and probability density forecast. Can i perform quantile regression with Gaussian ...
1
vote
0
answers
11
views
Problem formulation of future timeframe prediction based on current time
I have a problem where I want to predict "when is the next action happening" based on the time.
Example problem: Imagine you have a dataset of transactions per user, your goal is to predict ...
- 31
0
votes
0
answers
19
views
How do i quantitatively evaluate Prediction Interval accuracy using Gaussian Process Regression?
I am using a time series data using GPR Model and then want to quantitatively evaluated Prediction Intervals accuracy with PICP (PI coverage probability) and PINAW (PI normalized average width) for ...
2
votes
0
answers
142
views
How to assess calibration of probability distribution for a multiclass model?
I have a multiclass classifier (boosting model), and my goal is to have a good approximation of the actual distribution to the classes given my feature values.
I.e. suppose I have features $X$, and ...
- 56
2
votes
0
answers
92
views
One-vs-rest vs one-vs-one multiclass probability validation: does it matter?
Now that I have figured out how rms::val.prob works to the extent that I have written my own Python implementation, I would like to extend that idea to multiple ...
- 51.2k
0
votes
0
answers
83
views
How to interpret training and validation loss of DeepAR?
Please bear with me. Its a long but complete post.
My questions are:
Why does the training loss start to osccilate wildly after some epochs? It is because it has jumped out of a local minima? I tried ...
- 223
0
votes
0
answers
79
views
understanding what pattern DeepAR learns and how it works
I am trying to understand how the following simple pattern from a synthetic count time series is learnt using DeepAR. https://arxiv.org/abs/1704.04110
The count time series is generated from a ...
- 223
2
votes
1
answer
40
views
Should Bayesian probability of model given the data coincide with intuitive estimation of this probability?
I have $N$ real valued targets ordered by one single real valued feature. I would like to build a simple probabilistic model that states that all the targets corresponding to the features that are ...
- 5,350
0
votes
1
answer
63
views
Scale-Invariant CRPS Alternative
I am currently working on a probabilistic forecasting problem (outputting the full predictive distribution, possibly in the form of samples) and I need to decide on a measure to evaluate the forecasts....
- 54.5k
1
vote
0
answers
130
views
How can I generate probabilistic forecasts to do probabilistic classification?
I have a collection of univariate, irregularly spaced, financial time series. Each series is labeled by its class. The image below shows some example data.
A note on the data:
The time series could ...
- 124
1
vote
2
answers
119
views
Predictive distribution in the Frequentist setting
Let $\{X_t\}_{t \in T}$ be a time series, such that $X \sim F(\theta)$, for some arbitrary distribution $F$. Based on the observed values $\{x_1, x_2, \cdots, x_{t-1}\}$, suppose that I want to ...
- 111k
1
vote
3
answers
98
views
How to combine state-level COVID-19 vaccination rates with national demographic data at the individual level?
I’m investigating possible correlations between the COVID-19 vaccination rate in the United States and the results of a long-running survey of scientific personality traits like "Agreeableness&...
- 1,063
0
votes
0
answers
77
views
Computing probabilistic forecasts for a ratio of two other given forecasts
I have three time series (Revenue, Volume and Price), with Price = Revenue / Volume.
I have created probabilistic forecasts for Revenue and Volume and now need to derive a consistent forecast (...
- 289
3
votes
1
answer
159
views
Nonlinear regression: What's the best estimator for this problem?
Context and goals:
Consider $x$ a scalar (deterministic) independent variable and $y = f(x;\beta)+\eta$ be a dependent random variable obtained trough $f$ with some parameters $\beta$ and $\eta$ is ...
- 553
0
votes
0
answers
98
views
Which algorithm suits for classification of multivariate time series data?
I have multivariate time series data, where the goal is to predict a binary label, which is changing in time. For illustration: there are 20 individuals, for each of them I measure 50 values (...
- 71.6k
2
votes
0
answers
368
views
What method would I use when data has an upper and lower boundary in GluonTS
I have a time series dataset that has a max ceiling of 50 and a lower threshold of 0, however when using probabilistic time series modeling like GluonTS I am getting future predictions that are either ...
- 1,011
2
votes
1
answer
31
views
Distribution that acts like Poisson/NegBin for small means and like a Normal distribution for large means?
I want to generate a full density probabilistic forecasting model, where I don't know a priori whether the time series I want to model are intermittent or dense. In both cases, the time series is a ...
- 1,351