Missing observations in the dynamic regression model

By Franz C. Palm, T. Nijman. 1984

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Causal Discovery from Temporally Aggregated Time Series (2017, background): “If the noise terms follow the Gaussian distribution, A is usually not identifiable (Palm & Nijman, 1984).”
AN INDICATOR OF MONTHLY GDP AND AN EARLY ESTIMATE OF QUARTERLY GDP GROWTH (2005, unknown): “This approach was developed from the early work of Friedman (1962) by Chow and Lin (1971), Ginsburgh (1973) and Fernandez (1981) with parameter estimation addressed by Palm and Nijman (1984).”
Temporal aggregation of univariate and multivariate time series models: a survey (2008, unknown): “The whole subject is made more complex by parameter identification issues, see Palm and Nijman (1984).”
Temporal Disaggregation By State Space Methods: Dynamic Regression Methods Revisited (2006, unknown): “The behaviour of the large sample likelihood function of the Litterman model can be investigated using the approach by Palm and Nijman (1984). Lettingzt denote the sum of s consecutive observations, zt = S(L)yt, with S(L) = 1 + L + L(2) + · · ·Ls−1, originating from the ARIMA(1,1,0) process ∆yt = φ∆”
Discovering Temporal Causal Relations from Subsampled Data (2015, background): “It has been shown in (Palm & Nijman, 1984; Harvey, 1989) that with only the second-order information of the low-resolution data, usually the temporal causal relations are not identifiable.”
Diagnostic Checking in ARMA Models With Uncorrelated Errors (2005, methodology): “The aggregation properties of ARMA models are well known (see, e.g., Amemiya and Wu 1972; Harvey and Pierse 1984; Palm and Nijman 1984).”
Causal Discovery and Inference: Concepts and Recent Methodological Advances (2016, background): “Traditionally, if one uses only the second-order information, this suffers from parameter identification issues (Palm and Nijman 1984), i.e., the same subsampled (low-frequency) model may disaggregate to several high frequency models, which are observationally equivalent at the low frequency.”