In this paper a system for recommending investment channel choices to members of defined contribution retirement funds is proposed. The system is interactive, using a member's answers to a series of questions to derive a utility function. The observed values are interpolated by means of appropriate formulae to produce a smooth utility function over the whole positive range of benefits at retirement. The resulting function, together with stochastic models of the returns on the available channels and of the annuity factor at exit, is then used to recommend an optimum apportionment of the member's investment. The proposed system is applied to the observed values of utility functions of post-retirement income elicited from members of retirement funds. Difficulties in the application are discussed and the results are analysed. The sensitivity of the recommendations to the parameters of the stochastic model is discussed.

No abstract

The response of winter wheat to surface waterlogging in monoliths of clay soil in lysimeters was studied. The treatments, selected as corresponding to the most extreme likely in Britain, were: 5 days of waterlogging in October between germination and emergence; 42 days of waterlogging during tillering in January and February; and 21 days from mid-April during stem elongation. These treatments were used singly and in all combinations and were compared with a freely-drained control. During waterlogging, the oxygen concentration in the soil declined most rapidly when the soil was warmer (in October and April) and at the shallower depths (5, 20 cm); also, a decrease in soil solution nitrate was accompanied by an increase in nitrous oxide evolution. Pre-emergence waterlogging restricted plant population to 35 m-* compared to 338 m-2 under free drainage. Subsequently, compensatory growth resulted from diminished interplant competition for nitrogen, water and light; a greater proportion of tillers survived and the flag leaves were larger, although ears at harvest were fewer in all treatments waterlogged before emergence. After waterlogging during tillering, grain loss was accounted for entirely by a decrease in the number of ear-bearing tillers. After waterlogging during stem elongation, yield losses were associated primarily with fewer grains per spikelet. Grain yield was depressed by about 4 1 2 % after single waterloggings, by 20% after any two waterloggings, and by 32% when plants were waterlogged on all three occasions.

The effects of waterlogging, for up to 8 days at different stages of development, on the growth and yield of peas have been examined. The peas were grown outdoors subject to field conditions in lysimeters containing monolith sandy loam or clay soils, in facilities where drainage and rainfall could be controlled. The experiments, made during 2 years, involved waterlogging treatments likely to be the most extreme occurring in the pea-growing areas of the UK. During each waterlogging treatment the oxygen concentration at 20 cm depth in the soil declined to less than 2 within 2 days. On average, when the soil was waterlogged to the surface for 5 days at the 1-2, 3-4 and 6-7 leaf stages (the latter being just before flowering), the losses of yield of seeds (peas) at the freezing stage were 6, 15 and 42%, respectively. The severity of the effect depended on the duration of the treatment and the height of the water-table;at the 6-7 leaf stage either 2 days waterlogging to the surface or 5 days with the water-table a t 50 cm below the soil surface had little effect on seed yield. Waterlogging restricted leaf expansion and internode extension. After waterlogging at the pre-flowering stage, leaves usually senesced prematurely and the shoot apex became prematurely quiescent. In all experiments the yield of haulm was more affected by waterlogging than were the yields of pods and seeds. The losses in yield were mainly a consequence of fewer pods per plant. Waterlogging before flowering for 5 days depressed the concentrations of nitrogen, phosphorus and potassium in the haulm grown on both soils and of nitrogen in the seeds (peas) of plants in 1 year on the clay. Other treatments had little effect on nutrient concentration. After waterlogging t o the surface for 5 days before flowering the uptake of water by roots was restricted, but other treatments had only a small effect, being least in the early treatments.

No abstract