Long Memory in US Real Output per Capita

Question

1. What are the contributions in "Long memory in us real output per capita" ?

2. What have the authors stated for future works in "Long memory in us real output per capita" ?

3. What is the main advantage of the semi-parametric methods?

4. What is the importance of a robustness check?

Accepted Answer

This paper analyses the long memory properties of quarterly real output per capita in the US ( 1948Q1 – 2008Q3 ) using non-parametric, semi-parametric and parametric techniques.. The authors also examine the possibility of a structural break in the data and the results indicate that there is a slight reduction in the degree of persistence after the break that is found to occur in the second quarter of 1978.

Accepted Answer

The authors also examine the possibility of a structural break in the data and the results indicate that there is a slight reduction in the degree of persistence after the break that is found to occur in the second quarter of 1978.

Accepted Answer

The main advantage of the semi-parametric methods is that they only specify the I(d) structure, without modelling the d-differenced process, which is merely described as an I(0) process.

Accepted Answer

It is well known that the findings on persistence and long memory can vary substantially depending on the method used, and therefore a robustness check is crucial.

Accepted Answer

The choice of the bandwidth is crucial in view of the trade-off between bias and variance: the asymptotic variance is decreasing with m while the bias is growing with m.14per capita GDP, and the authors found some evidence of long memory (d > 0) in the growth rate series if the bandwidth is large, a result that might be spurious and reflect the neglected ARMA structure for the d-differenced process.

Accepted Answer

This is the standard case of I(d) models of the form: ,...,1,0,)1( tuxL tt d (1) ,0,0 txt where L is the lag operator (Lxt = xt-1) and ut is I(0) defined as a covariance stationary process with spectral density function that is positive and bounded at any frequency.

Accepted Answer

7In this section the authors briefly describe the methods employed for the empirical analysis in Section 4, which are based on non-parametric, semi-parametric and parametric techniques for modelling long-range dependence.

Accepted Answer

a limitation of the above approaches is that they do not allow the inclusion of deterministic terms such as intercepts and linear trends.

Accepted Answer

According to Giraitis et al. (2003) the V/S test is more suitable for series that exhibit high volatility, and various Monte Carlo experiments conducted by these authors show that the V/S test is less sensitive to the choice of the bandwidth number q.

Accepted Answer

Another advantage of this method is that it is valid for any real value of d, therefore encompassing stationary (d < 0.5) and nonstationary (d ≥ 0.5) hypotheses, unlike the methods described above that require first differencing to render the series stationary prior to the estimation of d.

Accepted Answer

Other methods such as Ohanissian et al. (2008) can also be employed, though this method exploits the invariance property of the long-memory parameter for temporal aggregation.

Accepted Answer

the process ut could itself be a stationary and invertible ARMA sequence, when its autocovariances decay exponentially; however, it could decay at a much slower rate than exponentially.

Accepted Answer

As for the choice of the optimal bandwidth parameter q (q*) in the two methods described above, q = 0 corresponds to the classic Hurst-Mandelbrot R/S statistic.

Accepted Answer

for example, if it is I(0) stationary, shocks affecting the series will be transitory and the degree of decay will depend on the structure describing the short-run dynamics, being, for example, exponential if they are autoregressive.

Long Memory in US Real Output per Capita

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What are the contributions in "Long memory in us real output per capita" ?

2. What have the authors stated for future works in "Long memory in us real output per capita" ?

3. What is the main advantage of the semi-parametric methods?

4. What is the importance of a robustness check?

5. What is the significance of the bandwidth?

6. What is the standard case of I(d) models of the form?

7. What are the methods used for the empirical analysis in Section 4?

8. What are the limitations of the above approaches?

9. What is the way to test the V/S test?

10. what is the advantage of the method?

11. What other methods can be used to estimate the degree of persistence after a break?

12. What is the simplest explanation for the ARMA sequence?

13. What is the way to determine the optimal bandwidth?

14. What is the difference between a stationary and an exponential ARMA sequence?

Figures

Citations

Weak and Strong Cross Section Dependence and Estimation of Large Panels

Testing of Unit Root and Other Nonstationary Hypotheses in Macroeconomic Time Series - (Now published in 'Journal of Econometrics', 80, 1997, pp.241-268.)

Enforcement of Contribution Norms in Public Good Games with Heterogeneous Populations

Fiscal Decentralization and Economic Growth in OECD Countries: Is There a Relationship?

Business Models for Media Firms: Does Competition Matter for How They Raise Revenue?

References

Time Series Tests of Endogenous Growth Models

Long memory relationships and the aggregation of dynamic models

Gaussian Semiparametric Estimation of Long Range Dependence

Empirical cross-section dynamics in economic growth

The Classical Approach to Convergence Analysis

Related Papers (5)

Business Strategy and Firm Reorganization under Changing Market Conditions

The Effects of School Competition on Academic Achievement and Grading Standards

Capital Structure and Regulation in U.S. Local Telephony: An Exploratory Econometric Study

One Size Fits All? Decentralization, Corruption, and the Monitoring of Bureaucrats

An Experimental Investigation of Colonel Blotto Games