Abstract
Non-stationary time series are a frequently observed phenomenon in several applied fields, particularly physics, engineering and economics. The conventional way of analysing such series has been via stationarity inducing filters. This can interfere with the intrinsic features of the series and induce distortions in the spectrum. To avert this possibility, it might be a better alternative to proceed directly with the series via the so-called time-varying spectrum. This article outlines the circumstances under which such an approach is possible, drawing attention to the practical applicability of these methods. Several methods are discussed and their relative advantages and drawbacks delineated.
Similar content being viewed by others
Notes
These and related concepts are discussed at length in Nachane (2006), chapters 3 and 4.
The concept of Lebesgue measure can be found in any standard text on real analysis such as e.g. Royden and Fitzpatrick (2010). A measure υ is said to be absolutely continuous w.r.t. a measure μ if for every measurable set A, with μ(A) = 0, we have υ(A) = 0.
Although the evolutionary spectrum defined by (51) is not invariant to the choice of the family ℑ, the integral \(\int_{ - \infty }^{\infty } {{\text{d}}H_{t} (\omega )\, = \,\text{var} [X(t)]}\) is independent of this choice.
The problem of resolution in spectral analysis refers to the ability of estimators to distinguish fine structure in the spectrum. The resolution of an estimator can be improved by decreasing its bandwidth [see Koopmans (1995), p. 303–306].
The term sufficiently apart means that either (i) \(\left| {\lambda_{1} \pm \lambda_{2} } \right| > > {\text{bandwidth}}\;{\text{of}} \left| {{\Gamma}({\lambda})} \right|^{2}\) or (ii) \(\left| {t_{1} - t_{2} } \right| > > {\text{width}}\;{\text{of}} \left\{ {w_{q} } \right\}.\).
Cumulants are defined and discussed in Brillinger (1975), p. 19–21.
The notion of weak dependence is discussed in Billingsley (1968), p. 363–367.
The notions of small o and big O are explained in Nachane (2006), p. 131.
The suggested test has reasonable size properties and the power of the test increases substantially with sample size. Other properties of the test are noted in Breitung and Candelon (2006).
The degrees of freedom are however adjusted to F[2, N − 3p].
The term BRICS is now a common acronym for the following group of countries—Brazil, Russia, India, China and S. Africa.
References
Akin, C., & Kose, M. A. (2008). Changing nature of north-south linkages: Stylized facts and explanations. Journal of Asian Economics, 19(1), 1–28.
Bentkus, R. J., & Zurbenko, I. G. (1976). Asymptotic normality of spectral estimates Dokaldi Akademic Nauk. SSSR, 229(1), 11–14. (in Russian).
Billingsley, P. (1968). Probability and measure. New York: Wiley.
Bradley, R. C. (2005). Basic properties of strong mixing conditions: A survey and some open questions. Probability Surveys, 2, 107–144.
Breitung, J., & Candelon, B. (2006). Testing for short- and long-run causality: A frequency domain approach. Journal of Econometrics, 12, 363–378.
Brillinger, D. R. (1975). Time series: Data analysis and theory, holt. New York: Rinehart & Winston.
Chan, W. Y. T., & Tong, H. (1975). A simulation study of the estimation of evolutionary spectral functions. Applied Statistics, 24, 333–341.
Chen, W., Khetarnavaz, N., & Spencer, T. W. (1993). An efficient algorithm for time-varying Fourier transform. Proc IEEE, 41, 2488–2490.
Cramer, H. (1961). On some classes of nonstationary stochastic processes. In Proceedings of the 4th Berkeley Symposium, vol. 2, p. 57–78
Dolado, J., & Lutkepohl, H. (1996). Making Wald tests work for cointegrated VAR systems. Econometric Reviews, 15, 369–386.
Doukhan, P. (1994). Mixing: Properties and examples. Lecture notes in statistics (Vol. 85). Berlin: Springer.
Doukhan, P., & Louhichi, S. (1999). A new weak dependence condition and applications to moment inequalities. Stochastic Processes and their Applications, 84(2), 313–342.
Falk, M. (1984). On the convergence of spectral densities of arrays of weakly stationary processes. Annals of Probability, 12, 918–921.
Fidrmuc, J., Korhonen, I., & Btorov, I. (2008). China in the World Economy: Dynamic Correlation Analysis of Business Cycles, BOFIT Discussion Paper No. 7/2008, Bank of Finland
Fishman, G. S. (1969). Spectral methods in econometrics. Cambridge: Harvard University Press.
Geweke, J. (1982). Measurement of linear dependence and feedback between multiple time series. Journal of the American Statistical Association, 77(378), 304–313.
Geweke, J. (1984). Measures of conditional linear dependence and feedback between time series. Journal of the American Statistical Association, 79(388), 907–915.
Granger, C. W. J. (1969). Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 37(3), 251–276.
Grenander, U. (1958). Resolvability and reliability in spectral analysis. Journal of Royal Statistical Society Series B, 20, 152–157.
Grenander, U., & Rosenblatt, M. (1957). Statistical analysis of stationary time series. New York: Wiley.
Hannan, E. J. (1967). Time series analysis, science paperbacks. London: Chapman & Hall.
Herbst, L. J. (1964). Spectral analysis in the presence of variance fluctuations. Journal of the Royal Statistical Society Series B, 2, 354–360.
Hlawatsch, F., & Boudreaux-Bartels, G. F. (1992). Linear and quadratic time–frequency signal representations. IEEE Signal Processing Magazine, 9(2), 21–67.
Ibragimov, I. A. (1962). Stationary Gaussian sequences that satisfy the strong mixing conditions Dokaldi Akademic Nauk. SSSR, 147(6), 1282–1284.
Ibragimov, I. A., & Linnic, Y. C. (1971). Independent and stationary sequences of random variables. Groningen: Volters-Nordoff.
Iosifescu, M. (1977). Limit theorems for φ-mixing sequences: A survey. In Proceedings of the fifth conference on probability theory, Brasov (Romania), Publishing House of the Romanian Academy
Ivanov, A. V., & Leonenko, N. N. (1989). Statistical analysis of random fields. Boston: Kluwer.
Karhunen, K. (1947). Uber lineare methoden in der Wahrscheinlichkeitsrechnung. Annals of the Academy of Sciences Finland, 37, 1–79.
Kay, S. M. (1989). Modern spectrum analysis. Englewood Cliffs: Prentice-Hall.
Kolmogorov, A. N., & Rozanov, Yu A. (1960). On strong mixing conditions for stationary Gaussian processes. Theory of Probability and Applications, 5(2), 204–208.
Kolmogorov, A.N. & Zurbenko, I.G. (1978). Estimation of spectral functions of stochastic processes. In Paper presented at the 11th European meeting of statisticians, Oslo
Koopmans, L. H. (1995) The spectral analysis of time series. San Diego, California: Academic Press.
Kose, M.A., Otrok, C., & Prasad, E.S. (2008). Global business cycles: Convergence or decoupling? IMF Working Paper No. WP/08/143
Loynes, R. M. (1968). On the concept of the spectrum for nonstationary processes. Journal of the Royal Statistical Society Series B, 30, 1–30.
Mark, W. D. (1970). Spectral analysis of the convolution and filtering of nonstationary stochastic processes. Journal of Sound Vibrations, 11, 19.
Martin, W., & Flandrin, P. (1985). Wigner-Ville spectral analysis of non-stationary processes. IEEE Transactions ASSP, 33, 1460–1470.
Melard, G. (1978). Proprieties du spectra evolutif d’un processus nonstationnnaire. Annales de l’Institut Henri Poincare, 14, 411–424.
Melard, G. (1985). An example of the evolutionary spectrum theory. Journal of Time Series Analysis, 6(2), 81–90.
Melard, G., & Schutter, A. (1989). Contributions to evolutionary spectral theory. Journal of Time Series Analysis, 10(1), 41–63.
Meyer, Y. (1993). Wavelets: Algorithms and applications. Philadelphia: SIAM.
Moran, P. A. P. (1953). The statistical analysis of the Canadian lynx cycle I: Structure and prediction. Australian Journal of Zoology, 1, 163–173.
Morrison, D. F. (1976). Multivariate statistical methods. New York: McGraw-Hill.
Nachane, D. M. (1997). Purchasing power parity: An analysis based on the evolutionary spectrum. Applied Economics, 29, 1515–1524.
Nachane, D. M. (2006). Econometrics: Theoretical foundations and empirical perspectives. Delhi: Oxford University Press.
Nachane, D. M., & Dubey, A. K. (2013) Trend and cyclical decoupling: New estimates based on spectral causality tests and wavelet correlations. Applied Economics, 45(31), 4419–4428.
Nachane, D. M., & Ray, D. (1993). Modelling exchange rate dynamics: New perspectives from the frequency domain. Journal of Forecasting, 12, 379–394.
Nagabhushanam, K., & Bhagwan, C. S. K. (1968). Non-stationary processes and spectrum. Canadian Journal of Mathematics, 20, 1203–1206.
Naidu, P. S. (1996). Modern spectrum analysis of time series. New York: CRC Press.
Nikias, C. L., & Petropulu, A. P. (1993). Higher-order spectral analysis: A nonlinear signal processing framework. New York: Prentice-Hall Inc.
Page, C. H. (1952). Instantaneous power spectra. Journal of Applied Physics, 23, 1203–1206.
Parzen, E. (1957). On consistent estimates of the spectrum of a stationary time series. Annals of Mathematical Statistics, 28, 329–348.
Parzen, E. (1967). On empirical multiple time series analysis. In Proceedings of the 5th Berkeley symposium on mathematical statistics & probability, p. 305–340
Percival, B. P., & Walden, A. T. (1998). Spectral analysis for physical applications: Multitaper and conventional univariate techniques. Cambridge: Cambridge University Press.
Portnoff, M. R. (1980). Time–frequency representations of digital signals and systems based on short-time Fourier analysis. IEEE Transactions ASSP, 28, 55–69.
Priestley, M. B. (1965). Evolutionary spectra and non-stationary processes. Journal of the Royal Statistical Society Series B, 27, 204–237.
Priestley, M. B. (1966). Design relations for non-stationary processes. Journal of the Royal Statistical Society Series B, 28, 228–240.
Priestley, M. B. (1969). Estimation of transfer functions in closed loop stochastic systems. Automatica, 5, 623–632.
Priestley, M. B. (1981). Spectral analysis and time series. London: Academic Press.
Priestley, M. B. (1988). Non-linear and non-stationary time series analysis. London: Academic Press.
Priestley, M. B., & Rao, T. S. (1969). A test for stationarity of time series. Journal of the Royal Statistical Society Series B, 31, 140–149.
Priestley, M. B., & Tong, H. (1973). On the analysis of bivariate non-stationary processes. Journal of the Royal Statistical Society Series B, 35, 153–166.
Rosenblatt, M. (1985). Stationary sequences and random fields. Boston: Birkhauser.
Royden, H. L., & Fitzpatrick, P. M. (2010). Real analysis (4th ed.). Noida: Pearson.
Slutzky, E. (1937). The summation of random causes as the source of cyclical processes. Econometrica, 5, 105–146.
Sun, M., Li, C., Sekhar, L. N., & Sclabassi, R. J. (1989). Efficient computation of the discrete Pseudo-Wigner distribution. IEEE Transactions ASSP, 37, 1735–1741.
Tjostheim, D. (1976). Spectral generating operators for non-stationary processes. Advances in Applied Probability, 8, 831–846.
Toda, H. Y., & Phillps, P. C. B. (1993). Vector autoregressions and causality. Econometrica, 61(6), 1367–1393.
Toda, H. Y., & Yamamoto, T. (1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 98, 225–255.
Vorobjev, L. S., & Zurbenko, I. G. (1979). The bounds for the power of C(α)-tests and their applications. Teoriya Veroyatnostei i ee Primeneniya, 24(2), 252–266. (in Russian).
Welch, P. D. (1967). The use of fast Fourier transform for estimation of power spectra: A method based on time-averaging over short, modified periodograms. IEEE Transactions AU, 15, 70–73.
Wigner, E. (1932). On the quantum correction for thermodynamic equilibrium. Physics Reviews, 40, 749–759.
Yao, F., & Hosoya, Y. (2000). Inference on one-way effect and evidence on Japanese macroeconomic data. Journal of Econometrics, 98, 225–255.
Zurbenko, I. G. (1978). On a statistic for the spectral density of a stationary sequence Dokaldi Akademic Nauk. SSSR, 239(1), 34–37.
Zurbenko, I. G. (1980). On effectiveness of estimations of the spectral density of a stationary process. Teoriya Veroyatnostei i ee Primeneniya, 25(3), 476–489. (in Russian).
Zurbenko, I. G. (1982). On consistent estimators for higher spectral densities Dokaldi Akademic Nauk. SSSR, 264(3), 529–532.
Zurbenko, I. G. (1986). The spectral analysis of time series. Amsterdam: North-Holland.
Zurbenko, I. G. (1991). Spectral analysis of non-stationary time series. International Statistical Review, 59, 163–174.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nachane, D.M. Time-varying spectral analysis: theory and applications. Ind. Econ. Rev. 53, 3–27 (2018). https://doi.org/10.1007/s41775-018-0030-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s41775-018-0030-2