Questions tagged [stochastic-processes]
A stochastic process describes the evolution of random variables/systems over time and/or space and/or any other index set. It has applications in areas such as econometrics, weather, signal processing, etc. Examples: Gaussian process, Markov process, etc.
1,280
questions
0
votes
0
answers
17
views
What is the cubic expectation (third-order moment) of a complex gaussian vector (say, E[$aa^{T}a$])?
Note: I also posted this question on MATHEMATICS.
For a real gaussian vector, an explicit formula for the cubic expectation can be found in Matrix Reference Manual (search 'Cubic Expectations' in this ...
1
vote
1
answer
61
views
What's the transition probability according to this PDE?
I'm trying to figure out how I can simulate markov chains based on an ODE:
dN/dt = alpha N (1 - N / K) - beta N
Thus N denotes ...
- 113
1
vote
0
answers
14
views
Moving Average with non-increasing coefficients
Suppose I have a Moving Average $MA(q)$.
$$X_t = \sum_{j=0}^q \psi_j \epsilon_{t-j}, \quad (\epsilon_t)_{t\in \mathbb{Z}}\,\,\ i.i.d.$$
I know that there is no imposition regarding the monoticity of ...
- 877
1
vote
1
answer
46
views
How to simulate non-gaussian stochastic paths
(Edited to be clearer)
I am trying to replicate simulating Geometric Brownian Motion (GBM) but instead of the stochastic increment following a normal distribution, I would like it to follow a ...
- 13
0
votes
0
answers
6
views
Variance of random walks in time series analysis [duplicate]
“For a random walk stochastic process, the variance is infinite.” Do you agree? Why?
0
votes
0
answers
26
views
About ergodic stochastic process
$x_t=2.1+0.73x_{t-1}+ε_t$
$ε_t \sim iid(0,σ_ε^2)$
Given the stochastic process with deep Gaussian process above, is x_t an ergodic stochastic process? If possible, I would like to know the reason.
- 1
0
votes
0
answers
10
views
The gamble problem to find limiting stationary distribution
[This question was asked here before, but I did not get satisfactory response]
I was trying to solve the following problem:
I find out the transition matrix is:
$$P = \begin{bmatrix}
1 & 0 &0 ...
- 243
1
vote
0
answers
13
views
Time series that are correlated in levels but not first differences
Is it correct to say that the two time series are correlated in levels, but not first differences?
$$x_{t}=\phi t+u_{t}\\ y_{t}=\rho t+\epsilon_{t}$$
where $\epsilon_{t}$ and $u_{t}$ are mutually ...
- 131
0
votes
0
answers
40
views
How does covariance stationarity even exist?
I've been wondering recently about covariance stationarity. Say we have a stationary series with statsmodels' ADF and KPSS results:
...
- 121
1
vote
1
answer
35
views
Predicting $x_t$ knowing something about $\Delta^2 x_t$
I have this exercise question, what is your prediction of $x_{10}$ knowing that $\Delta^2 x_t = \epsilon_t $, knowing $x_9 = 1.56$ and $x_8 = 1.64$.
I take this to mean that
$$
x_t = x_{t-2} +\...
- 511
0
votes
0
answers
31
views
Alternative to SimPy for continuous event simulation?
The python library, SimPy, is pretty explicit that it only handles discrete event simulation.
Though it is theoretically possible to do continuous simulations with SimPy, it has no features that help ...
- 1,700
0
votes
0
answers
11
views
Numerically solving the Forward equation to estimate SDEs
In books [1] dealing with inference for SDEs, why is the approach of numerically solving the forward PDE to obtain numerical estimate of the PTD not studied? One could then use this PTD to perform a ...
- 101
2
votes
2
answers
48
views
Stationarity and ergodicity of a process conditional on a finite trajectory
Let us say we are interested in a single time series, e.g. the daily closing share price of Tesla. We can model it as a realization of a stochastic process $\{Y_t(\omega)\}$. It corresponds to a ...
- 55.5k
0
votes
0
answers
15
views
First order PDF of a stochastic process
I've started studying about stochastic processes and I need some help in this question.
A random number generator is making numbers by this process:
First number (X0) is a sample from Normal Standard ...
- 1
4
votes
2
answers
71
views
Stochastic modelling, distribution and ergodicity of a particular time series with a given finite history
Let $\Omega$ be a sample space. A stochastic process $\{Y_t\}$ is a function of both time $t \in \{1, 2, 3, \ldots\}$ and outcome $\omega \in \Omega$.
For any time $t$, $Y_t$ is a random variable (i....
- 55.5k
1
vote
0
answers
30
views
Question about autocorrelation of time series
I have a time series $S(t)$ for $t\in I $ where $I=[0,t_\text{max}]$ is an interval.
The time series is very regular, $S(t) = \sin(t)$, although my following question is independent of the explicit ...
- 213
2
votes
1
answer
74
views
What is wrong with using t-SNE for predictions?
If I have a dataset with hundreds of samples and thousands of features, and t-SNE does a good job of separating classes compared to others classifiers, I don't understand why I can't rerun the ...
- 81
0
votes
0
answers
10
views
How to change some parameters of a process to make it WSS?
Consider the R.P., $X(t)=Au(t-T)$ where,
$$A \sim N(\mu,\sigma^2)$$
$$T \sim Exp(\lambda)$$
and $u(t)$ is the unit step function. If $A$ and $T$ are independent we'll have,
$$E(X(t))=\mu(1-\exp\{-\...
0
votes
0
answers
9
views
Application of Time reversibility property of stochastic processes
Are there any theoretical or applied consequences of a stochastic process being time reversible? I know a Markov chain being time-reversible implies the existence of a stationary distribution. But in ...
- 101
1
vote
1
answer
45
views
Proving the expectation of a variable in a stochastic process
Problem
Information packets arrive at a server with a poisson process having rate $\lambda = 2$ per hour.
The server processing time for a packet follows the distribution : $f(x) = 1, 0\leq x\leq1$
...
- 23
0
votes
0
answers
21
views
Plim of time average of a stochastic process, across independent realisations
I have a stochastic process $\{X_t\}_{t\in\mathbb{z}}$. Further, I have $N$ independent realisations of this process. What guarantees that $plim_{T\rightarrow\infty} \frac{1}{T}\sum_{t=1}^{\infty}X_{t}...
- 131
0
votes
1
answer
156
views
Some help on this question about stochastic epidemic models without removals would be greatly appreciated
I get this question up until the underlined part, would someone be able to explain the rest of it to me as its had me stumped for a little while now.
- 11
3
votes
1
answer
176
views
Is this statement of a stationary density function correct?
I'm planning to use a discrete-time stochastic process defined in the following paper:
Nicolau, J. (2002). Stationary Processes That Look Like Random Walks—The Bounded Random Walk Process in Discrete ...
- 155
1
vote
0
answers
20
views
If $(X_t)$ is a Gaussian process, what is $Cov(X_s^2,X_t^2)$?
Let $(X_t)_{t \in \mathbb R^+}$ be a Gaussian process $\mathcal N(\mu(t), \sigma^2(t))$ with covariance function $K$.
Let $s \leq t$. Can we express $Cov(X_s^2,X_t^2)$ in terms of $K, \mu(t), \sigma^2(...
- 163
0
votes
0
answers
17
views
How do you consider the negative states i.e. ...-3,-2,-1, when solving for the stationary distribution of this MC?
I have shown that the MC is aperiodic and irreducible, if that helps.
- 1
0
votes
0
answers
33
views
How can one find a system of SDE's from a probability density function?
Suppose I have a joint distribution function say $p(x,y,z)=f_{X, Y, Z}(x,y,z)$. Is it possible to find a system of stochastic differential equations or a single stochastic differential equation from ...
- 135
1
vote
0
answers
18
views
Sample path boundedness of Gaussian Processes
Consider a Gaussian Process $GP(0,k(x,x'))$ with zero mean and bounded, continues covariance function $k(x,x')<c,\quad \forall x,x\in\mathbb{R}^n$.
Are the sample paths of this process (almost ...
- 41
0
votes
1
answer
36
views
Show that two random vectors have the same distribution
I'm unsure of a result that looks simple, but I want to make sure I'm not getting it wrong.
Let $(\epsilon_t, t \in \mathbb{Z})$ be a process such that $\epsilon_t$ i.i.d.. I want to show that for any ...
- 223
0
votes
1
answer
40
views
Renewal process: Stationary and independent increments?
For a renewal process, we know that the inter-arrivals are independent but not exponentially distributed, as opposed to the Poisson process for which the inter-arrivals are exponential.
We also know ...
- 133
1
vote
0
answers
100
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
4
votes
1
answer
213
views
Meaning of "$\stackrel{p}\longrightarrow$" in math notation (arrow with a p over it)
I have a problem with the concept of a symbol (an arrow with a P over it). Can anyone clarify this for me? For example what the symbol means in this relationship:
$$\large Y_t^{(n)} \stackrel{p}\...
- 53
0
votes
0
answers
26
views
A nondeterministic covariance-stationary process approximated by an ARMA process
We know that the Wold Decomposition Theorem says that any purely nondeterministic covariance-stationary process, $x = [x_t : t \in \mathbb{Z}]$, can be written as a linear combination of lagged values ...
- 877
0
votes
1
answer
71
views
Winning at tennis
What is your probability of winning a game of tennis, starting from the even score Deuce(40-40), if your probability of winning each point is 0.3 and your opponent's is 0.7?
My answer:
I think the ...
- 886
1
vote
0
answers
54
views
Modeling Urns and Balls System as a Markov Chain
Suppose I have $q$ urns each of which hold up to $n$ distinguishable balls, but only $1$ of each type of ball (there being $n$ types of balls too). I would like to make any particular configuration of ...
- 151
2
votes
1
answer
56
views
Show that two white noise definitions are equivalent
Given $(\varepsilon_t)\sim WN(0,\sigma^2)$ a white noise. By definition
$$E(\varepsilon_t)=0,\,\, E(\varepsilon_t^2)=\sigma^2 \quad \forall t$$
and
$$E(\varepsilon_t \varepsilon_s) = 0, \quad s\neq t$$...
- 877
3
votes
0
answers
42
views
A conceptual question about the limitation of the MA processes
We know that linear time-series techniques are frequently used in macroeconometrics. The Wold Representation Theorem states that any covariance-stationary process may be expressed as an MA process ...
- 877
0
votes
0
answers
16
views
Coefficients for higher dimensional NIPC (non intrusive polynomial chaos expansion) heavily biased towards 0 degree function
Currently, I am trying to apply the PCE to the thermal fin problem as solved using finite element. Although I cannot attach the code I use, suffice to say I solve a linear system $AU=F$ where $F$ is ...
- 111
0
votes
0
answers
35
views
Simulating paths of stochastic process from density
I need yout help! I have a stochastic process $X_t$ and I know its density function $f(x,t)$, which is defined for $x>t$.
I'm looking for a code in R that simulates the paths of the process, so I ...
- 1
1
vote
0
answers
13
views
What would be a continuous-time version of a VAR process?
It is often said that a AR(1) process can be viewed as a discretized version of the continuous-time Ornstein-Uhlenbeck process. Can we really claim this to be valid considering that the Ornstein-...
- 299
1
vote
1
answer
26
views
Purpose of negative indices in time series
In the book Time-series analysis by Hamilton we find the following passage:
A time series is a collection of observations indexed by the date of each observation. Usually we have collected data ...
- 107
3
votes
1
answer
43
views
Why is the difference between 2 time series drawn from the same process not White Noise?
I take the difference between 2 time series (each with 200,000 observations) drawn from the same ARMA(2,1) process and find that (at least the first 1000 observations of) this difference looks like ...
- 3,242
0
votes
0
answers
17
views
Derivation of autocovariances of squared shocks
I have two shocks: $\varepsilon_{1t}$ has constant volatility $E[\varepsilon_{1t}^2]$ = $\sigma^2_1$ while $\varepsilon_{2t}$ has time varying volatility $E[\varepsilon_{2t}^2]$ = $\sigma^2_{2,t}$. I ...
- 137
0
votes
0
answers
26
views
The sum of $n$ zero-mean values is of $O(\sqrt{n})$ at max
Why is the sum of $n$ zero-mean values is of $O(\sqrt{n})$ at max?
If $X$ is a random variable with a standard deviation $\sigma$, the standard deviation of $n$ such i.i.d. random variables is $\sqrt{...
- 261
0
votes
0
answers
26
views
White Random Process(White Noise) and Mean Value
Can someone give a definition of a white random process?
I was trying to understand the penultimate paragraph of this answer
Is it necessary for white noise to have zero mean
which states that there ...
- 11
-4
votes
1
answer
68
views
Wait so probabilities of 0 or 1 CAN change? $P(A|B)=1$ does not imply $P(A)=1$ because $0 < P(A=B) < 1$?
Nassim Nicholas Taleb says here
no probability that is 0 or 1 should ever change.
Despite these 6 questions
Does an unconditional probability of 1 or 0 imply a conditional probability of 1 or 0 if ...
- 2,176
0
votes
0
answers
36
views
Can we conclude that Stochastic Gradient Descent is a "superior" algorithm that Gradient Descent?
On a very informal level, if we were to compare the (classical) Gradient Descent Algorithm to the Stochastic Gradient Descent Algorithm, the first thing that comes to mind is:
Gradient Descent can be ...
- 5,698
1
vote
0
answers
9
views
Does there exist a bounded measure for a certain stochastic process which follows a non-semigroup algebraic structure?
It is known that if we have a Markov process $(X_t)_{t \geq 0}$, then for all bounded measurable $f$ and for all $s,t \in \mathbb{R}$, there exists a bounded measure $P_sf$ such that
$$
E[f(X_{t + s}) ...
0
votes
0
answers
37
views
What is the expectation of a variable that has two cases (being either 0 or a constant) depending on a normal distributed variable
I need to determine the expectation of a variable: $\mathbb{E}(b(v))$
The value $b$ has two cases: it is either 0 when $v<0$ or a constant c when $v\geq 0$. The value $v$ has a normal distribution.
...
- 29
0
votes
0
answers
36
views
Proof of Stochastic Gradient Descent
When it comes to the classical Gradient Descent Algorithm, for optimizing Convex Functions - there is a standard proof that demonstrates the convergence of this algorithm provided certain conditions ...
- 5,698
1
vote
0
answers
24
views
Understanding the notation of a MA process in a paper
I am trying to understand a equation in a paper that I am reading.
First, considere de $MA(\infty)$ process definition:
\begin{equation}
(I)\quad \quad \quad X_t = \sum_{j=0}^\infty \psi_j \...
- 877