Abstract
In classical autoregressive models, it is assumed that the disturbances are normally distributed and the exogenous variable is non-stochastic. However, in practice, short-tailed symmetric disturbances occur frequently and exogenous variable is actually stochastic. In this paper, estimation of the parameters in autoregressive models with stochastic exogenous variable and non-normal disturbances both having short-tailed symmetric distribution is considered. This is the first study in this area as known to the authors. In this situation, maximum likelihood estimation technique is problematic and requires numerical solution which may have convergence problems and can cause bias. Besides, statistical properties of the estimators can not be obtained due to non-explicit functions. It is also known that least squares estimation technique yields neither efficient nor robust estimators. Therefore, modified maximum likelihood estimation technique is utilized in this study. It is shown that the estimators are highly efficient, robust to plausible alternatives having different forms of symmetric short-tailedness in the sample and explicit functions of data overcoming the necessity of numerical solution. A real life application is also given.
Similar content being viewed by others
References
Akkaya AD, Tiku ML (2001) Estimating parameters in autoregressive models in non-normal situations: asymmetric innovations. Commun Stat Theory Methods 30:517–536
Akkaya AD, Tiku ML (2005a) Robust estimation and hypothesis testing under short-tailedness and inliers. Test 14(1):129–150
Akkaya AD, Tiku ML (2005b) Time series AR(1) model for short-tailed distributions. Statistics 39:117–132
Akkaya AD, Tiku ML (2008a) Short-tailed distributions and inliers. Test 17:282–296
Akkaya AD, Tiku ML (2008b) Autoregressive models with short-tailed symmetric distributions. Statistics 42:207–221
Bass FM, Clarke DG (1972) Testing distributed lag models of advertising effect. J Mark Res 9:298–308
Bayrak OT, Akkaya AD (2010) Estimating parameters of a multiple autoregressive model by the modified maximum likelihood method. J Comput Appl Math 233:1763–1772
Bayrak OT, Akkaya AD (2011) Autoregressive models with stochastic design variables and nonnormal innovations. In: Recent researches in applied mathematics, simulation and modeling, proceedings of the 5th international conference on applied mathematics, simulation, modeling (ASM’11), Corfu Island, Greece, pp 197–201. ISSN:1792-4332
Box GEP, Jenkins GM, Reinsel GC (2008) Time series analysis, forecasting and control. Wiley, New Jersey
Hand DJ, Daly F, Lunn AD, McConway KJ, Ostrowski E (1994) Small data sets. Chapman and Hall, New York
Hoeffding W (1953) On the distribution of the expected value of the order statistics. Ann Math Stat 24:93–100
Islam MQ, Tiku ML (2004) Multiple linear regression model under nonnormality. Commun Stat Theory Methods 33:2443–2467
Islam MQ, Tiku ML (2010) Multiple linear regression model with stochastic design variables. J Appl Stat 37:923–943
Joiner BL, Rosenblatt JR (1971) Some properties of the range in samples from Tukey’s symmetric lambda distributions. J Am Stat Assoc 66:394–400
Kendall MG, Stuart A (1979) The advanced theory of statistics, vol 2. Charles-Griffin, London
Pearson ES, Tiku ML (1970) Some notes on the relationship between the distributions of central and non-central F. Biometrika 57:175–179
Puthenpura S, Sinha NK (1986) Modified maximum likelihood method for the robust estimation of system parameters from very noisy data. Automatica 22:231–235
Qumsiyeh SB (2007) Non-normal bivariate distributions: estimation and hypothesis testing. Dissertation, Middle East Technical University
Sazak HS, Tiku ML, Islam MQ (2006) Regression analysis with a stochastic design variable. Int Stat Rev 74:77–88
Schneider H (1986) Truncated and censored samples from normal populations. Marcel Dekker, New York
Thode HC (2002) Testing for normality. Marcel Dekker, New York
Tiku ML (1967) Estimating the mean and standard deviation from a censored normal sample. Biometrika 54:155–165
Tiku ML (1968) Estimating the parameters of normal and logistic distributions from censored samples. Aust J Stat 10:64–74
Tiku ML (1988) Order statistics in goodness-of-fit tests. Commun Stat Theory Methods 17:2369–2387
Tiku ML, Akkaya AD (2004) Robust estimation and hypothesis testing. New Age International Publishers/Oscar Publications, New Delhi
Tiku ML, Suresh RP (1992) A new method of estimation for location and scale parameters. J Stat Plan Inference 30:281–292
Tiku ML, Vaughan DC (1997) Logistic and nonlogistic density functions in binary regression with nonstochastic covariates. Biom J 39:883–898
Tiku ML, Vaughan DC (1999) A family of short-tailed symmetric distributions. Technical report, McMaster University
Tiku ML, Tan WY, Balakrishnan N (1986) Robust inference. Marcel Dekker, New York
Tiku ML, Wong WK, Vaughan DC, Bian G (2000) Time series models in non-normal situations: symmetric innovations. J Time Ser Anal 21:571–596
Vaughan DC (2002) The generalized secant hyperbolic distribution and its properties. Commun Stat Theory Methods 31:219–238
Vaughan DC, Tiku ML (2000) Estimation and hypothesis testing for a non-normal bivariate distribution with applications. J Math Comput Model 32:53–67
Yee TW, Huang K (n.d.) Tikuv {VGAM}. Retrieved from http://finzi.psych.upenn.edu/R/library/VGAM/html/tikuvUC.html
Acknowledgements
We sincerely thank the Chief Editor Dr. Ahmad Sheykhi and the referees for their invaluable comments.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
Proof of asymptotic equivalence of ML and MML
If f(z) is the probability density function of a random variable z, and z (i) (1 ≤ i ≤ n) are the order statistics of a random sample of size n with t (i) = E{z (i)} all finite, then (Hoeffding 1953)
where g(z) is a function which satisfies the condition |g(z)| ≤ h(z), h(z) being convex. Since the nonlinear functions \(h(u_{i} ) = u_{i} /(1 + [\lambda_{2} /2r_{2} ]u_{i}^{2} )\) and \(g(z_{i} ) = z_{i} /(1 + [\lambda_{1} /2r_{1} ]z_{i}^{2} ),\) given in Eq. (6) are all bounded, Eq. (19) is satisfied yielding the asymptotic equivalence of modified likelihood equations and likelihood equations.
Appendix 2
2.1 Sample Information Matrix
The sample information matrix is − 1 times the second derivatives of ln L evaluated at \(\mu_{1} = \hat{\mu }_{1} ,\) \(\sigma_{1} = \hat{\sigma }_{1} ,\) \(\gamma_{1} = \hat{\gamma }_{1} ,\) etc. Asymptotic variances and covariances are obtained from the inverse of this matrix. They provide accurate approximations; see for example the [100,000/n] Monte Carlo run results given in Table 4, where \(\gamma_{0} = 1.0,\) \(\gamma_{1} = 1.5,\) \(\phi = 0.5,\) \(\mu_{1} = 0.0\) and \(\sigma_{1}\) and \(\sigma\) are 1.0. As can be seen the simulated variances given in Table 1 and the ones obtained from sample information matrix are close to each other.
Appendix 3
3.1 Asymptotic Variances of the Related Parameters
Write \(w_{1i}^{*} = 1 - \lambda_{1} \{ 1 - (\lambda_{1} /2r_{1} )z_{i}^{2} \} /\{ 1 + (\lambda_{1} /2r_{1} )z_{i}^{2} \}^{2}\) and z i are iid and have the same distribution as z = ε/σ in Eq. (3).
since \(\frac{1}{{\sqrt {2\pi } }}\int_{ - \infty }^{\infty } {z^{2j} } \exp \left\{ { - \frac{{z^{2} }}{2}} \right\}\,{\text{d}}z = \frac{(2j)!}{{2^{j} (j)!}}\) and
It may be noted that \(w_{1i}^{*}\) is essentially an increasing function of \(z_{i}^{2}\) and is bounded. For large n, \(E( z_{(i)} ) \cong t_{(i)} .\) Since variance of \(z_{(i)}\) tends to zero as n tends to infinity, the following results (asymptotic) are obtained:
and since u i and z i are independent of each other and complete sums are invariant to ordering
Similarly,
and
where \(\bar{y}_{[i] - 1} = \frac{1}{{m_{1} }}\sum\nolimits_{i = 1}^{n} {\beta_{1i} y_{[i] - 1} } .\)
More specifically,
since − 1 < ϕ < 1 so that Y is stationary.
Rights and permissions
About this article
Cite this article
Bayrak, Ö.T., Akkaya, A.D. Inference of Autoregressive Model with Stochastic Exogenous Variable Under Short-Tailed Symmetric Distributions. Iran J Sci Technol Trans Sci 42, 2105–2116 (2018). https://doi.org/10.1007/s40995-017-0448-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40995-017-0448-x