Improvements in information technology have made it easier for industry to communicate with their customers, raising hopes for a scheme that can estimate when customers will want to make purchases. Although a number of models have been developed to estimate the time-varying purchase probability, they are based on very restrictive assumptions such as preceding purchase-event dependence and discrete-time effect of covariates. Our preliminary analysis of real-world data finds that these assumptions are invalid: self-exciting behavior, as well as marketing stimulus and preceding purchase dependence, should be examined as possible factors influencing purchase probability. In this paper, by employing the novel idea of hierarchical time rescaling, we propose a tractable but highly flexible model that can meld various types of intrinsic history dependency and marketing stimuli in a continuous-time setting. By employing the proposed model, which incorporates the three factors, we analyze actual data, and show that our model has the ability to precisely track the temporal dynamics of purchase probability at the level of individuals. It enables us to take effective marketing actions such as advertising and recommendations on timely and individual bases, leading to the construction of a profitable relationship with each customer.

The aim of this study was to evaluate the relationship between the serum levels of lipoprotein (a) [Lp (a)] and apolipoproteins (apo A‐1 and apo B) in schoolchildren with a history of coronary and cerebrovascular events in their grandparents. We measured serum concentrations of Lp (a) and apoliproteins immunochemically in 289 schoolchildren aged 12–13 years and questioned parents about coronary and cerebrovascular events in the children's grandparents. In boys and girls, mean ± s.d. levels of apo A‐1, apo B and Lp (a) were 134 ± 20.3 and 136 ± 17.4 mg/dL, 61 ± 16 and 66 ± 15 mg/dL and 12.5 ± 15.3 and 12.5 ± 15.1 mg/dL, respectively. There were no significant sex differences in the levels of apo A‐1, apo B, and Lp (a). The Lp (a) levels (mean ± s.d., 12.5 ± 15.2 mg/dL; median 7.5 mg/dL, n = 289) were not affected by other variables. The Lp (a) distribution was strongly positively skewed and 75% of schoolchildren had very low levels. In the total 289 schoolchildren, thirty‐two grandparents who had had coronary vascular events (21 myocardial infarction, 11 angina pectoris) and twenty‐three grandparents who had had cerebrovascular events were recorded. By the boxplot statistical analysis, no difference was found in Lp (a) levels in children whose grandparents had myocardial infarction compared with those whose grandparents had no such history, or compared with those whose grandparents had suffered cerebrovascular events. Analysis also showed that the values of log Lp (a) in children whose grandparents had myocardial infarction tended to be higher than the values in children whose grandparents had no such history (P = 0.09). No significant differences in the levels of apo A‐1 and apo B and in the apo B/A‐1 ratio could be seen between children grouped according to the presence or absence of coronary and cerebrovascular events in their grandparents. These results suggest that high levels of Lp (a) in schoolchildren aged 12–13 years may partly reflect the existence of coronary vascular disease in older family members. Lp (a) may account for the strongest index of family history to disease risk in comparison with other apolipoproteins. Further study is needed to clarify the appropriate mass measurement method for Lp (a) in schoolchildren.

Abstact To determine disturbances of lipid metabolism in patients with Kawasaki disease, we investigated changes in the serum levels of apolipoproteins and serum lipids. Results were as follows. Total serum cholesterol (TC) and high‐density lipoprotein cholesterol (HDL‐C) decreased during the early stage of Kawasaki disease. The apo A‐I and A‐II levels were low until the 2nd week of illness. Apo B decreased during the 1st week of illness, but rose slightly during the 2nd and 3rd weeks. The apo B/apo A‐I ratio and the apo B/TC ratio were high in the early stages of illness. The group of patients with coronary artery lesions showed low levels of TC and HDL‐C, and low levels of apo A‐I, A‐II and B during the early stages, compared with the group without coronary artery lesions. The apo B/TC ratio was significantly higher in the patients with coronary artery lesions during the 2nd week of illness. Our findings suggest an association between changes in serum lipids and apolipoproteins and coronary artery involvement in Kawasaki disease. These abnormalities may indicate the presence of early coronary arteriosclerosis.

The influence of obesity and fat distribution on serum levels of lipoprotein and apolipoprotein was investigated in 294 Japanese junior high school children (12‐13 years of age). Serum levels of low‐density lipoprotein cholesterol (LDLC) (P= 0.013), triglycerides (TG) (P= 0.0006), and apolipoprotein B (apoB) (P= 0.003), and the apoB/A‐I ratio (P= 0.005) were significantly higher and serum levels of high‐density lipoprotein cholesterol (HDLC) (P= 0.00003) and apoA‐1(P = 0.003) were significantly lower in obese boys than in non‐obese boys. The serum levels of TG (P = 0.013) and the apoB/A‐1 ratio (P= 0.011) were significantly higher and the serum levels of HDLC (P= 0.004) was significantly lower in obese girls than in non‐obese girls. The LDLC/apoB ratio was lower in obese girls than in non‐obese girls (P= 0.03). Obesity ( 20% of ideal weight) was strongly correlated with the serum levels of lipids and apolipoproteins in boys; this relationship was less clear in girls. The degree of obesity and the body mass index (BMI) were more strongly correlated with serum levels of lipids and apolipoproteins in boys than in girls. In boys, atherogenic‐lipoproteins and apolipoproteins, such as LDLC and apoB, showed a stronger correlation with the thickness of the triceps skinfold, while in girls the anti‐atherogenic lipoproteins and apolipoproteins, such as HDLC and apoA‐1, showed a stronger correlation with both the triceps and the subscapular skinfold thicknesses. In girls the relationships between the BMI and the degree of obesity and the thickness of the subscapular skinfold (S) thickness were similar to the relationships between those parameters and the triceps skinfold (T) thickness. In boys, these parameters showed a stronger correlation with the subscapular skinfold thickness than with the triceps skinfold thickness. The correlation coefficients for the relationships between skinfold thickness and lipid and apolipoprotein levels were similar to the coefficients for the relationships between skinfold thicknesses and the severity of obesity and the BMI. The distribution of central‐type fat accumulation, which is indicated by the thickness of the subscapular skinfold, the S/T ratio and S‐T value, was inversely correlated with the HDLC level in both boys and girls. The degree of obesity was strongly correlated with the atherogenic lipoprotein profile in boys, in part because the subscapular skinfold thickness was strongly correlated with the degree of obesity and the BMI. In girls, the correlations between indices of central‐type obesity and atherogenic lipid and apolipoprotein profiles were stronger than in boys. These data suggest that childhood obesity may be an early cardiovascular risk factor.

Because every spike of a neuron is determined by input signals, a train of spikes may contain information about the dynamics of unobserved neurons. A state-space method based on the leaky integrate-and-fire model, describing neuronal transformation from input signals to a spike train has been proposed for tracking input parameters represented by their mean and fluctuation [11]. In the present paper, we propose to make the estimation more realistic by adopting an LIF model augmented with an adaptive moving threshold. Moreover, because the direct state-space method is computationally infeasible for a data set comprising thousands of spikes, we further develop a practical method for transforming instantaneous firing characteristics back to input parameters. The instantaneous firing characteristics, represented by the firing rate and non-Poisson irregularity, can be estimated using a computationally feasible algorithm. We applied our proposed methods to synthetic data to clarify that they perform well.