题目1811 · 统计
Validity or Stationarity Check 1
Consider the proposal rho(h)=0.8^|h|. Is it valid from a stationarity / autocorrelation perspective?
打开 →找到 30 个结果
中文题目题目1812 · 统计
Validity or Stationarity Check 2
Consider the proposal rho(1)=1.2. Is it valid from a stationarity / autocorrelation perspective?
打开 →题目1813 · 统计
Validity or Stationarity Check 3
Consider the proposal An AR(1) with phi = 1.03. Is it valid from a stationarity / autocorrelation perspective?
打开 →题目1814 · 统计
Validity or Stationarity Check 4
Consider the proposal An AR(1) with phi = -0.6. Is it valid from a stationarity / autocorrelation perspective?
打开 →题目1815 · 统计
Validity or Stationarity Check 5
Consider the proposal An MA(1) claiming rho(1)=0.7. Is it valid from a stationarity / autocorrelation perspective?
打开 →题目6024 · 统计
Persistence and the Covariance-Stationarity Verdict
A GARCH(1,1) has $\alpha=0.20$, $\beta=0.75$. Compute the persistence $\alpha+\beta$ and state whether the process is covariance-stationary (i.e. has a finite, time-invariant unconditional variance). Give the persistence as a decimal.
打开 →题目1796 · 统计
Signal Stationarity Classification 1
A candidate signal is defined by X_t = ε_t + 0.3 ε_{t-1}. Is it weakly stationary?
打开 →题目1797 · 统计
Signal Stationarity Classification 2
A candidate signal is defined by X_t = 0.2 t + ε_t. Is it weakly stationary?
打开 →题目1798 · 统计
Signal Stationarity Classification 3
A candidate signal is defined by X_t = ε_t + s_t where s_t is a fixed deterministic day-of-week pattern. Is it weakly stationary?
打开 →题目1799 · 统计
Signal Stationarity Classification 4
A candidate signal is defined by X_t = A cos(ω t) + B sin(ω t), where E[A]=E[B]=0, Var(A)=Var(B), Cov(A,B)=0. Is it weakly stationary?
打开 →题目1800 · 统计
Signal Stationarity Classification 5
A candidate signal is defined by X_t = t ε_t. Is it weakly stationary?
打开 →题目4976 · 随机过程
Infer Missing Birth-Death Rate From Stationarity 11
A three-state birth-death CTMC has stationary distribution (0.5, 0.3, 0.2). The rates 0->1 and 1->2 are 0.6 and 0.4, and the rate 1->0 is 1. What rate 2->1 is needed for stationarity?
打开 →题目2905 · 概率
Ehrenfest Urn Stationarity
In the Ehrenfest urn model with $N$ balls, state $i$ means exactly $i$ balls are red. Each step, choose one ball uniformly at random and flip its color. Find the stationary distribution of the chain on $\{0,1,\dots,N\}$.
打开 →题目4979 · 随机过程
Infer Missing Birth-Death Rate From Stationarity 14
A three-state birth-death CTMC has stationary distribution (0.2, 0.5, 0.3). The rates 0->1 and 1->2 are 1.5 and 0.9, and the rate 1->0 is 0.6. What rate 2->1 is implied?
打开 →题目2902 · 概率
Laziness Does Not Change Stationarity
Suppose $\pi$ is stationary for a Markov chain with transition matrix $P$. Fix $\theta\in(0,1)$ and define a lazy version \[ P'=\theta I+(1-\theta)P. \] Show that $\pi$ is also stationary for $P'$.
打开 →题目4324 · 机器学习
Small Data With Local Stationarity
You have limited labeled data, and the target depends on local translation-equivariant patterns in a 2D signal map. Which architecture family usually brings the strongest built-in inductive bias?
打开 →题目2893 · 概率
Uniform Stationarity for a Doubly Stochastic Chain
Suppose a finite Markov chain has transition matrix $P$ whose rows and columns both sum to $1$. Show that the uniform distribution is stationary.
打开 →题目1816 · 统计
Why Differencing Helps a Trend but Not White Noise
A signal looks like deterministic trend plus short-memory noise. Why can first differencing help stationarity, while differencing plain white noise usually just injects a negative lag-1 correlation?
打开 →题目1818 · 统计
Why Ergodicity Matters
Why is ergodicity stronger than stationarity, and why do practitioners care about it when they average one long signal history?
打开 →模块2.3.1 · 数学与统计能力 · 时间序列分析
平稳性与 ARMA 模型
time-series · stationarity · autocorrelation · acf · pacf · white-noise · random-walk · ar
打开 →课程平稳性与 ARMA 模型 · 时间序列分析
平稳性与自相关函数
某私募(private fund)交易日下午四点,你的 PM 把过去 500 个交易日的策略净值推过来,问:这条曲线的均值真的稳定吗?波动率有没有结构性变化?只看一条路径,凭什么相信估出来的均值与自相关有意义?这是时间序列分析(time series analysis)的元问题。横截面统计里你有 公式 个独立同分布(i.i.d.)样本,推断建立在「重复抽样」...
打开 →题目2676 · 机器学习
A Tiny Tail Probability Can Dominate Average PnL
A strategy makes +0.04% on 98% of days and loses -2.5% on the remaining 2% of days. What is the unconditional average daily return?
打开 →课程平稳性与 ARMA 模型 · 时间序列分析
AR、MA 与 ARMA 过程
周一开盘前，某沪深300 量化私募的研究员把昨天打捞回来的 1500 个日内对数收益样本（log returns）丢进 R，画了一张样本 ACF：lag 1 大约 0.18，lag 2 大约 0.05，再往后几乎全部落进 Bartlett 带里。她想问的是：这条「拖尾」曲线像不像一阶自回归（autoregressive, AR）模型该有的样子？如果是 AR，...
打开 →题目1836 · 统计
AR(1) Forecast Error Variance 1
For the AR(1) model X_t = phi X_(t-1) + e_t with phi = 0.6 and Var(e_t) = 1, what is the h = 3 step forecast error variance?
打开 →题目1838 · 统计
AR(1) Forecast Error Variance 3
For the AR(1) model X_t = phi X_(t-1) + e_t with phi = 0.5 and Var(e_t) = 2.25, what is the h = 4 step forecast error variance?
打开 →题目1821 · 统计
AR(1) Multi-Step Forecast 1
A signal follows X_t = 0 + 0.6 X_(t-1) + e_t with Var(e_t) = 2 and current value X_t = 10. What is the h = 3 step forecast E[X_(t+3) | X_t]?
打开 →课程波动率与机制转换模型 · 时间序列分析
ARCH 与 GARCH 模型
某私募(private fund)的风控会上，研究员甩出沪深300 日收益的实证表：日内收益序列本身的自相关系数 公式 在滞后 公式 时几乎全部落在 公式 的 Bartlett 带内；可一旦把同一条序列 平方 再画一次 ACF，从滞后 1 到滞后 60 全是正值、缓慢衰减。再算样本峰度：5.8——远大于正态分布(Gaussian distributi...
打开 →题目1842 · 统计
ARMA Identification or Simplification 2
You observe the diagnostic statement: ACF cuts after lag 1, PACF tails off. What is the correct modeling conclusion?
打开 →题目1844 · 统计
ARMA Identification or Simplification 4
You observe the diagnostic statement: AIC prefers ARMA(2,1) but BIC prefers ARMA(1,1). What is the correct modeling conclusion?
打开 →课程平稳性与 ARMA 模型 · 时间序列分析
ARMA 模型的识别、估计与预测
周一早盘，某私募的时间序列研究员把过去 200 个交易日的对冲组合超额收益丢进 statsmodels。她想确认这条曲线是不是一个干净的 ARMA 过程——若是，残差就是一组白噪声，可以挂上下一阶段的 GARCH；若不是，她得回去重做特征工程。问题是：用 AR(1)、MA(1)、ARMA(1, 1) 还是 ARMA(2, 1)？拟合完之后怎么知道这一支模型确...
打开 →