1,356,188 research outputs found
Point and interval forecasts of mortality rates and life expectancy: A comparison of ten principal component methods
No full textUsing the age- and sex-specific data of 14 developed countries, we compare the point and interval forecast accuracy and bias of ten principal component methods for forecasting mortality rates and life expectancy. The ten methods are variants and extensions of the Lee-Carter method. Based on one-step forecast errors, the weighted Hyndman-Ullah method provides the most accurate point forecasts of mortality rates and the Lee-Miller method is the least biased. For the accuracy and bias of life expectancy, the weighted Hyndman-Ullah method performs the best for female mortality and the Lee-Miller method for male mortality. While all methods underestimate variability in mortality rates, the more complex Hyndman-Ullah methods are more accurate than the simpler methods. The weighted Hyndman-Ullah method provides the most accurate interval forecasts for mortality rates, while the robust Hyndman-Ullah method provides the best interval forecast accuracy for life expectancy.forecasting, forecasting time series, interval forecasts, Lee-Carter method, life expectancy, mortality forecasting, principal components analysis
Hyndman Peak
No full textA mountain is visible across a valley and between two hills. Description reads: ""Hyndman Peak (12,078 ft. elevation) as seen from upper Big Lost River near Kane Creek on Forest Road to Ketchum. Forest: Challis, State: Idaho, Date: 7/1940, Author: P.S. Bieler""
A comparison of ten principal component methods for forecasting mortality rates
No full textUsing the age- and sex-specific data of 14 developed countries, we compare the short- to medium-term accuracy of ten principal component methods for forecasting mortality rates and life expectancy. These ten methods include the Lee-Carter method and many of its variants and extensions. For forecasting mortality rates, the weighted Hyndman-Ullah method provides the most accurate point forecasts, while the Lee-Miller method gives the best point forecast accuracy of life expectancy. Furthermore, the weighted Hyndman-Ullah method provides the most accurate interval forecasts of mortality rates, while the robust Hyndman-Ullah method provides the best interval forecast accuracy of life expectancy.Mortality forecasting, life expectancy forecasting, principal component methods, Lee-Carter method, interval forecasts, forecasting time series
[Correspondence between Meyer Bodansky and James B. Hyndman - May-September 1940]
No full textLetters between Dr. Meyer Bodansky and James B. Hyndman, dated from May 16, 1940 to September 27, 1940. The letter that is dated May 16, 1940 is an invitation to Dr. Bodansky from Dr. Hyndman to be a guest speaker at the Texas Society of Medical Technologists annual Convention in October of the same year. The letters remaining discuss the details of the content of the lecture as well as the reimbursement of travel expenses. The last letter that is dated September 27, 1940 from Dr. Bodansky explains that he will not be able to attend the convention
"Stops walking when talking" as a predictor of falls in people with stroke living in the community
No full textOBJECTIVE: To test "Stops walking when talking" (SWWT) as a predictor of falls among people with stroke living in the community. METHODS: People with stroke were identified through hospital records. Mobility, ADL (activities of daily living) ability, mental state, mood, and SWWT were assessed in a single session. Participants were followed prospectively for six months, using falls diaries and regular telephone calls. RESULTS: Sixty three participants (36 men, 27 women mean (SD) age 68.4 (10.6)) were recruited. Four subjects had a brainstem lesion, 30 had right hemisphere, and 29 left hemisphere infarctions. Mean time since onset of stroke was 20 months (range 2-72). Twenty six subjects stopped walking when a conversation was started and 16 of them fell during the six month follow up period (11 experienced repeated falls). For all fallers (>or=1) the positive predictive value of SWWT was 62% (16/26), the negative predictive value 62% (23/37), specificity 70% (23/33) and sensitivity 53% (16/30). For repeat fallers (>or=2) the positive predictive value of SWWT was 42% (11/26), the negative predictive value 89% (33/37), specificity 69% (33/48) and sensitivity 73% (11/15). Those who stopped walking were significantly more disabled (p<0.001)-that is, they were more dependent in activities of daily living, had worse gross function as well as worse upper and lower limb function, and had depression (p = 0.012). CONCLUSIONS: The specificity of the SWWT test was lower but sensitivity was higher than previously reported. Although the SWWT test was easy to use, its clinical usefulness as a single indicator of fall risk in identifying those community dwelling people with stroke most at risk of falls and in need of therapeutic intervention is questionable
Fall events among people with stroke living in the community: Attention deficits, balance and falls (Proceedings of SRR)
No full textPURPOSE: To describe levels of attention deficits among people with stroke living in the community and explore relationships between attention, balance, function and falls. METHOD: Forty-eight mobile community-dwelling people with stroke (30 men, 18 women, mean age 68.4 +/- 11.2) were recruited to this cross-sectional investigation through General Practitioners. Twenty-six participants had a right, 21 a left hemisphere infarction and one had a brain stem lesion; mean time since stroke was 46 months (range five to 204). Participants' were interviewed about fall-events; attention, balance and function were assessed using standardised tests. RESULTS: Visual inattention was identified in five participants (10%), deficits of sustained attention in 15 (31%), auditory selective attention in nine (19%), visual selective attention in 17 (35%) and divided attention deficits in 21 participants (43%). Sustained and divided attention scores correlated with balance, ADL ability and fall-status (p < 0.01). The balance and function of subjects with normal attention were better than those with abnormal scores (p < 0.01). Analysis of variance revealed differences between repeat-fallers and non-fallers with no near-falls for divided attention, balance and ADL ability (p < 0.01). CONCLUSIONS: Attention deficits were common among this sample; sustained and divided attention deficits correlated with functional impairments and falls, highlighting that attention deficits might contribute to accident prone behaviour and falling
People with stroke living in the community: an investigation into the relationship between attention, functional ability and falls (abstract of paper in Proceedings of SSR)
No full textBackground: Information about the attention deficits of people with stroke living in the community is limited. The aims of this study were, to describe levels of attention and to explore the relationships between attention deficits, functional ability and fall events.
Method: Subjects living in the community were identified through GPs and therapy records. Assessments of balance, ADL function, attention and history of falls were completed in participants' homes for this cross-sectional study. Results: Forty-eight participants (30 men, 18 women mean age 68.4, SD 11.2) were recruited, 17 were repeat fallers, 7 single fallers, 12 were nonfallers with near falls and 12 were nonfallers with no near falls. One subject had a brainstem lesion, 26 had right and 21 left hemisphere infarctions. Mean time since onset of stroke was 46 months (range 5-204). Five (10.4%) participants presented with visual inattention, 15 (31%) had sustained attention deficits, 9 (19%) had auditory selective attention deficits, 17 (35%) had visual selective attention deficits and 21 (43%) presented with divided attention deficits. Sustained and divided attention scores were found to correlate with the scores for balance and ADL ability (p < 0.01). The balance and functional abilities of those subjects with normal attention scores were significantly better than those with abnormal scores (p < 0.01). Analysis of variance revealed significant differences between the fall groups for balance, ADL ability and divided attention; the greatest differences (p < 0.01) were between repeat fallers and nonfallers with no near falls.Conclusions: Attention deficits were common among community-dwelling people with stroke. Repeat fallers had significantly more problems dividing attention than nonfallers with no near falls. Those with impaired attention and those who had fallen repeatedly had significantly greater functional deficits
The vector innovation structural time series framework: a simple approach to multivariate forecasting
No full textThe vector innovation structural time series framework is proposed as a way of modelling a set of related time series. Like all multi-series approaches, the aim is to exploit potential inter-series dependencies to improve the fit and forecasts. A key feature of the framework is that the series are decomposed into common components such as trend and seasonal effects. Equations that describe the evolution of these components through time are used as the sole way of representing the inter-temporal dependencies. The approach is illustrated on a bivariate data set comprising Australian exchange rates of the UK pound and US dollar. Its forecasting capacity is compared to other common single- and multi-series approaches in an experiment using time series from a large macroeconomic database.Vector innovation structural time series, state space model, multivariate time series, exponential smoothing, forecast comparison, vector autoregression.
Nonparametric time series forecasting with dynamic updating
No full textWe present a nonparametric method to forecast a seasonal univariate time series, and propose four dynamic updating methods to improve point forecast accuracy. Our methods consider a seasonal univariate time series as a functional time series. We propose first to reduce the dimensionality by applying functional principal component analysis to the historical observations, and then to use univariate time series forecasting and functional principal component regression techniques. When data in the most recent year are partially observed, we improve point forecast accuracy using dynamic updating methods. We also introduce a nonparametric approach to construct prediction intervals of updated forecasts, and compare the empirical coverage probability with an existing parametric method. Our approaches are data-driven and computationally fast, and hence they are feasible to be applied in real time high frequency dynamic updating. The methods are demonstrated using monthly sea surface temperatures from 1950 to 2008.Functional time series, Functional principal component analysis, Ordinary least squares, Penalized least squares, Ridge regression, Sea surface temperatures, Seasonal time series.
Hierarchical forecasts for Australian domestic tourism
No full textIn this paper we explore the hierarchical nature of tourism demand time series and produce short-term forecasts for Australian domestic tourism. The data and forecasts are organized in a hierarchy based on disaggregating the data for different geographical regions and for different purposes of travel. We consider five approaches to hierarchical forecasting: two variations of the top-down approach, the bottom-up method, a newly proposed top-down approach where top-level forecasts are disaggregated according to forecasted proportions of lower level series, and a recently proposed optimal combination approach. Our forecast performance evaluation shows that the top-down approach based on forecast proportions and the optimal combination method perform best for the tourism hierarchies we consider. By applying these methods, we produce detailed forecasts for the Australian domestic tourism market.Australia, exponential smoothing, hierarchical forecasting, innovations state space models, optimal combination forecasts, top-down method, tourism demand.
- …
