The　stock　price　of　a　given　company　is　not　only　affected　by　its　fundamentals　and　other　associated　listed　companies,　but　also　influenced　by　the　fluctuation　of　the　stock　market　index.　Network　analysis　methods　can　be　used　to　analysis　net　correlations　between　stocks.　This　paper　constructs　minimum　spanning　tree　(MST)　and　hierarchical　tree　(HT)　of　stocks　of　shanghai　sector　traded　in　Shanghai　Stock　Exchange　based　on　Partial　Correlation　Coefficients　between　stocks,　deleting　the　impacts　of　Shanghai　Composite　Index.　Then　the　distribution　features　of　stocks　are　analyzed.　Furthermore,　Partial　Correlation　Coefficients　and　Simple　Correlation　Coefficients　are　applied　to　constructing　MST　of　stocks,　the　empirical　results　are　compared　and　we　find　that　Partial　Correlation　Coefficient　is　a　better　choice　when　constructing　network　of　stocks.　©　2010　IEEE.