Newest Questions
219,221 questions
0
votes
0
answers
5
views
Is the term non-Markovian used in statistics?
This is something I am trying to find some references/reading materials on.
I have multiple states ($S_1$, $S_2$, $S_3$). I observe multiple observations over time for each subject.
I want to make a ...
- 73
3
votes
0
answers
22
views
How to quantify uncertainty for time series predictions
I'm conducting an interrupted time series analysis to estimate the effect of an intervention.
I fit a (ARIMA) time series model on pre-intervention data
I forecast 3 months post-intervention to ...
- 31
0
votes
0
answers
24
views
How would a statistician decide whether to go voting?
Guy the statistician wants to decide whether to go to vote or not. He will go to vote if and only if his vote will affect the outcome (in other words, he is a "pivotal voter"). His null ...
- 1,091
4
votes
2
answers
38
views
Finding Maximum Likelihood Estimator
Assume the data to be i.i.d. random variables $X_1, X_2, \dots, X_n$
with density $$f(x,p) = p e^{-x} + 2(1-p)e^{-2x}, \quad x \ge 0.$$ We
need to find the procedure of determining MLE of $p$ based ...
- 111
0
votes
1
answer
24
views
What is the significance of the Fisher information in modern ML and statistics?
Material I've seen on the Fisher information seems flush with "properties" and a bit lacking in cohesion. It's easy to catalogue properties of the Fisher information:
As the Hessian of the ...
- 363
0
votes
1
answer
22
views
Drawing sample if population parameters are known
If we know the parameters of the population, i.e., the mean, median, and standard deviation...
a) We know everything about this population, so there is no need to draw a sample
b) We can estimate in ...
0
votes
0
answers
22
views
How do I determine if my non-normally distributed results support my hypothesis? [closed]
Preface
This is a follow-up to my previous question about the same data, where there is extended discussion in the comments. The scope is a different and broader, as I have now realised that I do not ...
- 205
2
votes
0
answers
23
views
Modification of Markov processes
In a semi-Markov process:
The holding time in each state can follow a general distribution, but this distribution depends only on the current state.
Similarly, the transition probabilities to the ...
- 73
0
votes
0
answers
14
views
Discrepancy between delta Change and effect size values
We have conducted a study comparing two different training interventions. We used ANCOVA to analyse the data. Taking the example of countermovement jump (CMJ) test, findings indicated significant ...
- 1
3
votes
0
answers
20
views
Is the Fisher divergence obtained as the functional Bregman divergence applied to the Fisher information?
This specific question arose as I'm trying to understand the connection between Bregman divergences and the Fisher divergence in the context of discrete diffusion. While querying ChatGPT (yes...) for ...
- 363
2
votes
1
answer
32
views
Non-monotonic power for cumulative incidence in a single-arm trial
I am trying to calculate power for a single-arm trial with a competing risk endpoint. However, the power as a function of the sample size has a strange pattern. Please see below.
I do not think it is ...
0
votes
0
answers
16
views
Testing conditional (in)dependence of a binary variable on another binary and a count/continuous variable
Here both $x$ and $n$ can serve as biomarkers of a health condition $y$ (case/control). $x$ can take only two values, whereas $n$ is a count variable, which, if necessary, could be modeled as ...
- 5,111
0
votes
0
answers
39
views
What if AIC and f-test prefers one model but the coefficient estimates and standard errors are unstable?
I am modelling the influence of species proportions on a response. AIC and f-test both prefer the model with separate interactions between these species proportions, although the difference in AIC is ...
- 49
10
votes
3
answers
376
views
What are some ways to represent uncertainty on a map?
Hypothetical question, so I don't have actual data or visualization to share, but this is a problem I might face in the future.
Let's say I have a map of a region, divided by counties. I take samples ...
- 101
4
votes
1
answer
97
views
Definition of a tail-$\sigma$-algebra
The following is from Erhan Çinlar's "Probability and Stochastics"
Let $(\mathcal{G}_n)_n$ be a (not necessarily independent) sequence of sub-$\sigma$-algebras on $(\Omega,\mathcal{H})$. We ...
- 441