正在学习

Ullah, A.

Ullah, A.

unemployment, 12, 184, 200, 216, 236, 332, 490, 493, 499; predictability of future unemployment, 213–24; unemployment dynamics, 179. See also unemployment rate

unemployment rate, 15, 140, 146, 164, 171, 176, 177, 183, 220, 239, 241, 322, 324, 325, 362, 480; from 1970 to 2014, 212. See also three-variable system (CPI inflation rate, unemployment rate, three-month T-bill rate), empirical example of

univariate forecasting models, and trending variables, 470–72; long-horizon forecasts, 475–77, 476nn8–9; and out-of-sample (OoS) evaluation, 486–89; and risk for differing approaches to long-horizon forecasts, 477–78; short-horizon forecasts, 472–75

univariate Gaussian kernel, 246

univariate linear prediction models, 133–34, 166. See also autoregressive moving average (ARMA) models

univariate nonlinear prediction models, 166–67; testing for nonlinearity, 179–80. See also autoregressive (AR) models; nonlinear univariate models

University of Michigan, survey of U.S. consumers, 322

Vahid, F.

Valkanov, R., 481, 484, 485, 501, 502

van Dijk, D.

Varian, H. R.

VARMA models, 187n1, 218

vector autoregressions (VARs), 7, 117, 186, 212, 217, 218, 238, 408, 463, 478; and the choice of lag length, 190; constrained VARs, 189; estimation of VAR( p) models, 189–90; factor-augmented VAR (FAVAR) models, 8,

193–94, 232; first-order Markov switching VAR (MSVAR), 456; misspecification for, 192; multiperiod forecasts with, 191–93; OLS estimation of, 189–90, 191, 194, 197, 198, 201, 203, 221; popularity of, 214; Qual VAR, 268; specification of vector auto regressions, 186–88; use of to approximate covariance stationary multivariate processes, 186–87; VAR forecasts, 39. See also Bayesian VARs (BVARs)

Villani, M.

Violante, F.

Vogelsang, T. J.

volatility, and density forecasting, 275–77, 308–9, 427–28; forecasts using realized volatility measures, 288–90; role of the loss function in (optimal forecasts in first and second moments under different loss functions), 277–78

volatility models, 276, 278, 284, 284n4; estimation of parameters of location-scale models, 279–80; location-scale models, 278–80; stochastic volatility models, 199, 200, 287, 338, 459. See also generalized autoregressive conditional heteroscedasticity (GARCH) models; multivariate volatility models

volatility forecast comparisons

Waggoner, D. F.

Wahba, G.

Wald test

Waldmann, R.

Wallis, K. F.

Watson, M. W.

weak LNN results

weather forecasts/forecasters, 13, 38, 435n3

Weibull distribution

Weinbach, G. C.

Weiss, A. A.

Welch, I.

West, K. D. See also out-of-sample (OoS) average loss, conducting inference on

West, M.

White, H. See also Reality Check (White’s bootstrap Reality Check approach)

white noise, 135, 136–37, 138, 143, 227, 290, 359,359n5, 471, 508

Whiteman, C. H.

White’s QuickNet Algorithm

Wiener-Kolmogorov prediction formulas

Wilson, E. B.

Winkler, R. L.

Wishart distribution

Wohar, M. E.

Wolak, F. A.

Wold representation theorem, 133, 135, 140–41, 148, 186, 359

Wolf, M.

Woolridge, J. M.

Wouters, R.

Wright, J. H.

Wu, J. C.

Yang, Y.

Yogo, M.

Yoldas, E.

Yoo, B. S.

Zanakis, S. H.

Zarnowitz, V.

Zellner, A. exposition by of the Bayesian analysis of regression models, 85; g -prior approach of, 82, 459n13

Zha, T.

Zhang, H. H.

Zopounidis, C.

Zou, H.

练习题

What is the primary focus of Ullah's research on unemployment?

A. Predictability of future unemployment
B. Impact of technology on unemployment
C. Geographical distribution of unemployment
D. Historical trends in employment rates

Which period is specifically mentioned for the unemployment rate in the source material?

A. 1950 to 1970
B. 1970 to 2014
C. 1980 to 2000
D. 2000 to 2020

What type of forecasting models are discussed in relation to trending variables?

A. Multivariate forecasting models
B. Univariate forecasting models
C. Dynamic factor models
D. Panel data models

Which of the following are mentioned as components of vector autoregressions (VARs)?

A. Choice of lag length
B. Estimation of VAR(p) models
C. Use of to approximate covariance stationary multivariate processes
D. Application in weather forecasting
E. Multiperiod forecasts with VARs

Which of the following are types of volatility models mentioned in the source material?

A. Location-scale models
B. Stochastic volatility models
C. Generalized autoregressive conditional heteroscedasticity (GARCH) models
D. Vector autoregressions (VARs)
E. Multivariate volatility models

The Wold representation theorem is applicable to both univariate and multivariate time series analysis.

White noise is characterized by a constant mean and variance over time.

The factor-augmented VAR (FAVAR) models are mentioned as an extension of the basic VAR models, specifically in pages ___.

Explain the role of the loss function in volatility and density forecasting.

What is the significance of Zellner's g-prior approach in Bayesian analysis?

Which of the following statements about the Wold representation theorem are correct?

A. It applies to any stationary stochastic process.
B. It decomposes a stationary process into a deterministic part and a stochastic part.
C. It is only applicable to autoregressive processes.
D. It guarantees that any stationary process can be represented as an infinite moving average.
E. It is limited to finite-order moving average processes.

When estimating parameters of a vector autoregression (VAR) model, which tool can be used to construct the likelihood or pseudo-likelihood for the data, especially when the model contains unknown parameters in the coefficient matrices?

A. Kalman filter
B. Wold representation theorem
C. White noise test
D. Univariate Gaussian kernel

Which of the following statements are true regarding the properties and applications of vector autoregressions (VARs)? Select all that apply.

A. VARs can be used to approximate covariance stationary multivariate processes.
B. The choice of lag length is not important in VAR models.
C. VAR forecasts can be made using multiperiod forecasts.
D. The Wold representation theorem is not related to VARs.
E. VARs are popular for analyzing the dynamic relationship between multiple time series.

In a vector autoregression (VAR) model, the Kalman smoother is used to estimate the unobserved state variables at time given all past, current, and future data, while the Kalman filter is used to estimate these variables given only past and current data.

The ___ theorem states that any covariance stationary process can be represented as a linear combination of lagged values of a white noise process.

登录后解锁笔记、知识点解析、AI 问答

立即登录