In　the　constant-stress　partial　accelerated　life　test,　estimation　of　the　parameter　for　inverse　Rayleigh　distribution　is　studied　based　on　left　censored　samples.　Maximum　likelihood　method　is　used　to　obtain　the　maximum　likelihood　estimations　of　the　unknown　parameter　and　acceleration　factor.　The　Fisher　information　matrix　is　calculated　according　to　the　principle　of　missing　information　proposed　by　Louis,　meanwhile　the　approximate　confidence　intervals　of　the　distributed　parameter　and　the　acceleration　factor　are　obtained.　When　the　prior　distributions　of　the　parameters　are　exponential　distribution,　the　Bayesian　estimations　of　the　parameters　are　obtained　under　the　square　loss　function,　and　the　maximum　likelihood　method　is　used　to　get　the　estimations　of　the　super　parameters.　The　mean　square　errors　of　the　estimators　are　calculated　by　Monte　Carlo　simulation,　based　on　these,　the　maximum　likelihood　estimation　and　Bayesian　estimation　are　compared.　Finally,　the　parameter　of　the　inverse　Rayleigh　distribution　and　the　estimation　of　acceleration　factor　are　calculated　for　different　left　censored　samples.　©　2020,　Zhejiang　University　Press.　All　right　reserved.