钢铁价格走势预测研究

华中科技大学硕士学位论文钢铁价格走势预测研究姓名：刘晶申请学位级别：硕士专业：企业管理指导教师：田志龙20050420I20044IIAbstractWiththeintensityofcompetitioninsteelindustry,howtoexactlyforecastthesteel’spricebecomesanimportantissuethatshouldbepaidmoreattentionsbysteelcompanies.Thispaperfirstlydiscussestheindexofsteelindustryprosperityandanalyzesthesteelmarketstructuretounderstandthedevelopmenttrendofsteelindustry.Then,weforecastthesteel’spricefromtwoaspects,thatis,outsidedrivingfactorsanditsself-developingtrend.Outsidedrivingfactorsareanalyzedbyusingqualitativeandquantitativemethods,andtheoutcomescomingfromthesemethodscheckoutandsupporteachother.Intheanalyzingpriceself-developingtrend,thepaperadoptsfirstlyseveraltimeseriesmodelstoforecastthefutureprice,thensynthesizestheresultsbyusingacombinationmodeltoreachoptimization.Finally,thepapercontrastsandsynthesizestheforecastedoutcomeseducedfromthetwoaspectsofoutsidedrivingfactorsandself-developingtrendofprice,thatis,wesynthesizetheforecastedoutcomesoflong-termandshort-termdevelopmenttrendsandresultinthefinalforecastedpriceofsteelindustry.Thefinalresultisalsoexplainedinthepaper.Weconcludethefollowingconclusionsbythisstudy:WedoanempiricalstudyonthepriceofsteelindustryinApril2004byusingtheforecastsystemproposedbythispaper.Theresultsshowthatthepriceweforecastedisinaccordwiththepractice.ItmeansthatthemodelproposedbyusissuitableforforecastingsteelpriceinChina.Besides,bycontrastingvariousforecastingmodels,weconcludethatthemodelproposedbythispaperisscientificandbetter.Empiricalstudyalsoshowsitsoperationalizationofthemodel.Scientificandempiricalstudiesshowthatourmodelcanbeusedtoguidethesteelcompaniesinpricingandpromotetheircompetenceinprice.Keywords:SteelIndustryPriceForecastMarketStructureMultipleRegressionAnalysis_____111.11.1.11.1.2WTO21.21-11-131.31.3.1[1]1[2]HHIHHI=(Xi/Xi)2XiHHI4[3]1XiHHI=1/nnHHI0HHI=12HHIHHIHHI3n1/n0.12584HHIHHIHHI2--SCPq[4]q41.3.2202020[5]-[8][9]-[11][12]-[15]ARMAARMAARIMAARCHSETAR51.3.3[16][17]-[19][20]1[21][22]2[23][24]61.3.4[25]-[27]SSEMAEMSEMAPEMSPE[28][29]p-[30]-[33]71.41-2[34]1-21-27[35]-[39][40]81-11-1GDP922.12.1.1[41]2.1.220011200422-1100110120130140150160170180200123420022342003234200422-1=(Xi/Xi)2XiHHI41XiHHI=1/nnHHI0HHI=12HHIHHIHHI113n1/n0.12584HHIHHIHHI2HHI80%60199220032-2300350400450500550199219931994199519961997199819992000200120022003HHIHHI2-22-211992HHI,200337.7%2HHI6HHI18001000HHI1800500HHI1000200HHI500100HHI200HHI1002-219922.2.2112qq2199820042-3y=0.0272x-0.0509-5.00%0.00%5.00%10.00%15.00%20.00%19981999200020012002200320042-3=nn4.5.22004=113111353531113220042%6%3031520045.15-15-1325.25-1GDP*GDP+0.874*+0.672*+0.594*+0.534*+0.762*20.92*+0.932*+0.669*+0.627*+0.589*+0.68*30.502*+0.914*40.84*+0.657*+0.54150.975*+0.941*60.923*+0.909*70.834*+0.877*80.927*90.927*100.903*345.45.4.15-35-3LinearYb0b1X-QuadraticYb0b1Xb2X2Yb0b1X1X1X;Yb0b2X22X2X2CompoundYb0(b1X)ln(Y)ln(b0)[ln(b1)]XGrowthYe(b0b1X)ln(Y)b0b1XLogarithmicYb0b1ln(X)Yb0b1X1X1ln(X)ExponentialYb0e(b1X)ln(Y)ln(b0)b1XInverseYb0(b1X)Yb0b1X1X11XPowerYb0(Xb1)ln(Y)ln(b0)b1ln(X)355.4.28SPSSFFSPSS105-45-41012345678910Linear0.34520.25170.44020.50380.18930.52820.00020.5090.13630.7106Quadratic0.34540.59330.61680.84040.77660.72630.05820.56240.68370.7477Inverse0.24810.32610.81450.95730.050.78720.98960.69620.86020.98185-423467891015Inverse5.55.5.15.5