Knowledge of the seasonal variation in births and deaths during normal years is important for analyses of the effects of wars, famines, epidemics or similar privations on these variables. In studies of seasonality, multiple trigonometric regression models are more flexible than the simple sine curve. The seasonal variation in mortality in Iceland, 1856-1990, shows a strong secular decrease, and a connection between this and the epidemiological transition is considered. For equidistant data, the regressor vectors are orthogonal, resulting in parameter estimates that are uncorrelated and independent of the number of trigonometric terms in the models. Comparisons with findings in other European countries are made. The temporal trends in Iceland of the birth components; the twinning rate, the still birth rate and the secondary sex ratio, are presented and compared with the corresponding values in neighbouring countries. No marked differences were emerged. During the first stage of the epidemiological transition the seasonal variation in deaths and births was mainly a result of the living conditions. Therefore, years marked by severe famine or other crises have had strong effects on the seasonal patterns. This can be seen in the data from Iceland during the 19th century. For Icelandic males, one has to include drownings as an important indicator of the harsh living conditions. In the first half of the 19th century, the fishermen had a hard life when they practised their profession and braved the Atlantic storms in the small vessels of that time. Until the 20th century, the sex ratio for the Icelandic population was low relative to the sex ratio for both Norway and Denmark, indicating the greater effect of the fluctuations in mortality on males than on females.
Read full article: http://bp.bookpi.org/index.php/bpi/catalog/view/23/56/141-1