牛祥秋
摘 要 研究了如何確定離散時(shí)間情況下再保險(xiǎn)模型破產(chǎn)概率上界的問題.為了降低自身的破產(chǎn)風(fēng)險(xiǎn),保險(xiǎn)公司常常對(duì)部分乃至全部資產(chǎn)進(jìn)行再保險(xiǎn).假定索賠間隔時(shí)間和索賠額具有一階自回歸結(jié)構(gòu),假定利率過程為取值于可數(shù)狀態(tài)空間的Markov鏈.建立了其比例再保險(xiǎn)模型,分別用遞歸更新技巧和鞅方法得到模型的破產(chǎn)概率上界.該破產(chǎn)概率上界作為評(píng)估再保險(xiǎn)公司償付能力和風(fēng)險(xiǎn)控制能力的重要指標(biāo),對(duì)于它的研究成果能為再保險(xiǎn)人做出重大決策提供重要的依據(jù),具有較為重要的理論和現(xiàn)實(shí)意義.
關(guān)鍵詞 概率論; 上界; 鞅; 比例再保險(xiǎn); 破產(chǎn)概率; Markov鏈利率
中圖分類號(hào) O211.9 文獻(xiàn)標(biāo)識(shí)碼 A
Abstract Upper bounds for the ruin probability of reinsurance were studied in a discrete time risk model. To reduce the risk, there is a possibility to reinsure a part or the whole reserve. In the model, the time between the occurrence of the claims and the claims were assumed to be the AR(1) structure, the interest rates followed a Markov chain with a denumerable state space. The risk model of proportional reinsurance was considered. The upper bounds for the ruin probability were derived both by renewal recursive technique and martingale method. As an important indicator of the abilities of solvency and risk management, the research of the ruin probability can provide an important basis for reinsurer's major decisions, so it has important theoretical and practical significance.
Key words probability theory; upper bound ; Martingale; proportional reinsurance; Markov chain interest rate
1 引 言
破產(chǎn)概率是風(fēng)險(xiǎn)研究的內(nèi)容之一.保險(xiǎn)公司為了降低破產(chǎn)風(fēng)險(xiǎn)而傾向于把部分甚至是全部資產(chǎn)進(jìn)行再保險(xiǎn),因此對(duì)再保險(xiǎn)破產(chǎn)概率的研究很具有現(xiàn)實(shí)意義. Cai(2002a)研究了利率為獨(dú)立同分布的隨機(jī)變量情形的離散時(shí)間風(fēng)險(xiǎn)模型破產(chǎn)概率上界估計(jì)[1];Cai(2002b)研究了利率為一階自回歸情形的離散時(shí)間風(fēng)險(xiǎn)模型破產(chǎn)概率上界估計(jì)[2].Cai,Dickson(2004)考慮了利率為Markov鏈形式的風(fēng)險(xiǎn)模型,并且分別用遞歸方法和鞅方法得出了破產(chǎn)概率的上界[3]; Yang,Zhang(2003)研究了保費(fèi)和索賠額具有一階自回歸結(jié)構(gòu)的常利率風(fēng)險(xiǎn)模型,得到破產(chǎn)概率的指數(shù)型和非指數(shù)型上界[4]; Lin,Wang(2006)研究了凈損失額(索賠額減去保費(fèi))具有一階自回歸結(jié)構(gòu),利率為Markov鏈形式的風(fēng)險(xiǎn)模型,并且分別用歸納法和鞅方法得出了破產(chǎn)概率的上界[5];郭風(fēng)龍,王定成(2012)研究了保費(fèi)收入和利率均具有Markov鏈形式的風(fēng)險(xiǎn)模型,運(yùn)用遞歸更新方法得到破產(chǎn)概率的Lundberg型上界[6].魏龍飛(2016)研究了索賠和利率過程分別為2個(gè)自回歸移動(dòng)平均結(jié)構(gòu)模型的情形,運(yùn)用遞歸更新方法得到破產(chǎn)概率的上界估計(jì)并對(duì)兩類風(fēng)險(xiǎn)模型的破產(chǎn)概率的上界進(jìn)行了比較[7].程建華,王德輝(2012)研究了保費(fèi)和索賠額具有一階自回歸結(jié)構(gòu)而利率為Markov鏈形式的風(fēng)險(xiǎn)模型,針對(duì)保費(fèi)的期初收取和期末收取的兩種情況,分別用鞅方法得到其各自的破產(chǎn)概率上界[8].Diasparra,Romera(2009)研究了利率為Markov鏈形式而索賠間隔時(shí)間和索賠額均為獨(dú)立同分布的非負(fù)隨機(jī)變量序列的再保險(xiǎn)風(fēng)險(xiǎn)模型,同樣是用兩種方法得到破產(chǎn)概率的上界[9];基于文獻(xiàn)[9],王麗霞,李雙東(2014)運(yùn)用更新遞歸方法研究了利率具有一階自回歸結(jié)構(gòu)的再保險(xiǎn)模型的破產(chǎn)概率上界[10].是在文獻(xiàn)[9]模型的基礎(chǔ)上,考慮索賠間隔時(shí)間和索賠額具有一階自回歸結(jié)構(gòu)而利率為取值于可數(shù)狀態(tài)空間的齊次Markov鏈形式結(jié)構(gòu)的比例再保險(xiǎn)情形.運(yùn)用遞歸更新技巧得到比例再保險(xiǎn)模型的破產(chǎn)概率的微積分形式,并利用歸納法給出破產(chǎn)概率的Lundberg型上界;運(yùn)用鞅方法得出模型的破產(chǎn)概率的上界估計(jì).
5 結(jié) 論
隨著中國的保險(xiǎn)市場逐步與國際接軌,各保險(xiǎn)公司越來越重視到再保險(xiǎn)的重要性.保險(xiǎn)公司通過再保險(xiǎn)旨在分散和控制風(fēng)險(xiǎn)以達(dá)到降低破產(chǎn)發(fā)生的概率.而索賠間隔時(shí)間,索賠額,利率以及分保費(fèi)比例等因素直接影響到破產(chǎn)發(fā)生的概率.基于此,論文考慮了利率、索賠額和索賠的時(shí)間間隔的相依情形對(duì)比例再保險(xiǎn)模型破產(chǎn)概率的影響.分別運(yùn)用了更新遞歸技巧和鞅方法兩種方法得到模型的兩種破產(chǎn)概率的上界估計(jì).由此來分析各因素的變化對(duì)再保險(xiǎn)模型破產(chǎn)概率的影響,具有重要的現(xiàn)實(shí)意義.
對(duì)于再保險(xiǎn)的研究主要分為比例再保險(xiǎn)和超額損失再保險(xiǎn)兩個(gè)方向.文中僅考慮了比例再保險(xiǎn)的情形,未有涉及超額損失再保險(xiǎn)情形.接下來會(huì)對(duì)超額損失再保險(xiǎn)作進(jìn)一步研究.endprint
參考文獻(xiàn)
[1] J CAI. Discrete time risk models under rates of interest [J]. Probability in the engineering and informational sciences,2002,16(3):309-324.
[2] J CAI. Ruin probabilities with dependent rates of interest [J]. Journal of Applied Probability.2002,39(2): 312-323.
[3] J CAI, D DICKSON. Ruin probabilities with a Markov chain interest model [J]. Insurance Mathematics & Economics.2004,35(3):513-525.
[4] H L YANG, L H ZHANG. Martingale method for ruin Probability in an autoregressive model with constant interest Rate[J]. Probability in the Engineering & Informational Sciences,2003, 17(2):183-198.
[5] L XU, R M WANG. Upper bounds for ruin probabilities in an autoregressive risk model with a Markov chain interest rate [J].Journal lf industrial and management optimization,2006,2(2):165-175.
[6] 郭風(fēng)龍,王定成. 考慮Markov保費(fèi)和利率的離散時(shí) 間風(fēng)險(xiǎn)模型的破產(chǎn)概率[J].數(shù)學(xué)的實(shí)踐與認(rèn)識(shí), 2012, 42(12):136-140.
[7] 魏龍飛. 具有相依結(jié)構(gòu)離散時(shí)間模型破產(chǎn)概率的上界[J]. 經(jīng)濟(jì)數(shù)學(xué), 2016, 33(1):88-92.
[8] 程建華,王德輝. Markov鏈利率下相依風(fēng)險(xiǎn)模型破產(chǎn)概率的上界[J].吉林大學(xué)學(xué)報(bào):理學(xué)版,2012,50(2):173-178.
[9] M DIASPARRA, R ROMERA. Inequalities for the ruin probability in a controlled discrete-time risk process[J]. European Journal of Operational Research. 2009, 204(3): 496-504.
[10]王麗霞,李雙東.相依利率下離散時(shí)間再保險(xiǎn)模型的破產(chǎn)問題[J].應(yīng)用概率統(tǒng)計(jì),2014,30(3):279-288.
[11]D LAMBERTON, B LAPEYRE. Introduction to stochastic calculus applied to finance[M]. London: Chapman & Hall,1996.endprint