抽样调查总结

注:抽样期末考前随便写写的

 
 

性质简单随机抽样系统抽样分层抽样整群抽样不放回不等概抽样
总体均值 Y ˉ = 1 N ∑ i = 1 N Y i \bar Y = {1 \over N}\sum_{i=1}^NY_i Yˉ=N1i=1NYi Y ˉ = ∑ g = 1 G W g Y ˉ g \bar Y = \sum_{g=1}^GW_g\bar Y_g Yˉ=g=1GWgYˉg Y ˉ ˉ = 1 N ∑ i = 1 N 1 M ∑ j = 1 M Y i j \bar {\bar Y} = {1\over N}\sum_{i=1}^N{1\over M}\sum_{j=1}^MY_{ij} Yˉˉ=N1i=1NM1j=1MYij Y ^ H T \hat Y_{HT} Y^HT
总体总量 Y = ∑ i = 1 N Y i Y = \sum_{i=1}^NY_i Y=i=1NYi Y = ∑ g = 1 G Y g Y = \sum_{g=1}^GY_g Y=g=1GYg
总体方差 S 2 = 1 N − 1 ∑ i = 1 N ( Y i − Y ˉ ) 2 S^2 = {1\over N-1}\sum_{i=1}^N(Y_i-\bar Y)^2 S2=N11i=1N(YiYˉ)2 S 2 ≈ ∑ g = 1 G W g ( S g 2 + ( Y ˉ g − Y ˉ ) 2 ) S^2 \approx \sum_{g=1}^GW_g(S_g^2+(\bar Y_g - \bar Y)^2) S2g=1GWg(Sg2+(YˉgYˉ)2) S 2 = 1 N M − 1 ∑ i , j ( Y i j − Y ˉ ˉ ) 2 = S w 2 + S b 2 S^2 = {1\over NM-1}\sum_{i,j}(Y_{ij}-\bar {\bar Y})^2 = S_w^2 + S_b^2 S2=NM11i,j(YijYˉˉ)2=Sw2+Sb2
总体均值的估计 Y ˉ ^ = y ˉ \hat {\bar Y} = \bar y Yˉ^=yˉ y ˉ s r = 1 n ∑ i = 1 n y r j \bar y_{sr} = {1\over n}\sum_{i=1}^ny_{rj} yˉsr=n1i=1nyrj y ˉ s t = ∑ g = 1 G W g y ˉ g \bar y_{st} = \sum_{g=1}^GW_g\bar y_g yˉst=g=1GWgyˉg y ˉ ˉ = 1 n M ∑ i = 1 n ∑ j = 1 M y i j = y ˉ M \bar {\bar y} = {1\over nM}\sum_{i=1}^n\sum_{j=1}^My_{ij} = {\bar y \over M} yˉˉ=nM1i=1nj=1Myij=Myˉ
总体总量的估计 Y ^ = N y ˉ \hat Y = N\bar y Y^=Nyˉ Y ^ s t = N y ˉ s t \hat Y_{st} = N\bar y_{st} Y^st=Nyˉst
总体方差的估计 s 2 = 1 n − 1 ∑ i = 1 n ( y i − y ˉ ) 2 s^2 = {1\over n-1}\sum_{i=1}^n(y_i-\bar y)^2 s2=n11i=1n(yiyˉ)2
均值估计的期望 E ( y ˉ ) = Y ˉ E(\bar y) = \bar Y E(yˉ)=Yˉ E ( y ˉ s y ) = 1 k ∑ r = 1 k y ˉ r = Y ˉ , N = n k 时 E(\bar y_{sy}) = {1\over k}\sum_{r=1}^k\bar y_r = \bar Y, N = nk时 E(yˉsy)=k1r=1kyˉr=Yˉ,N=nk E ( y ˉ s t ) = Y ˉ E(\bar y_{st}) = \bar Y E(yˉst)=Yˉ
均值估计的方差 V ( y ˉ ) = ( 1 − f ) S 2 n V(\bar y) = (1-f){S^2 \over n} V(yˉ)=(1f)nS2 V ( y ˉ s y ) = 1 k ∑ r = 1 k ( y ˉ r − Y ˉ ) 2 V(\bar y_{sy}) = {1\over k}\sum_{r=1}^k(\bar y_r - \bar Y)^2 V(yˉsy)=k1r=1k(yˉrYˉ)2 V ( y ˉ s t ) = ∑ g = 1 G W g 2 ( 1 − n g N g ) S g 2 n g V(\bar y_{st}) = \sum_{g=1}^GW_g^2(1-{n_g \over N_g}){S_g^2 \over n_g} V(yˉst)=g=1GWg2(1Ngng)ngSg2 V ( y ˉ ˉ ) = V ( y ˉ M ) = 1 M 2 ( 1 − n N ) 1 n 1 N − 1 ∑ i = 1 N ( Y i − Y ˉ ) 2 V(\bar {\bar y}) = V({\bar y \over M}) = {1\over M^2}(1-{n\over N}){1\over n}{1\over N-1}\sum_{i=1}^N(Y_i-\bar Y)^2 V(yˉˉ)=V(Myˉ)=M21(1Nn)n1N11i=1N(YiYˉ)2
总量估计的期望 E ( N y ˉ ) = Y E(N\bar y) = Y E(Nyˉ)=Y E ( Y ^ s t ) = N E ( y ˉ s t ) = Y E(\hat Y_{st}) = NE(\bar y_{st}) = Y E(Y^st)=NE(yˉst)=Y
总量估计的方差 V ( N y ˉ ) = N ( N − n ) n S 2 V(N\bar y) = {N(N-n)\over n}S^2 V(Nyˉ)=nN(Nn)S2 V ( Y ^ s t ) = ∑ g = 1 G N g ( N g − n g ) S g 2 n g V(\hat Y_{st}) = \sum_{g=1}^GN_g(N_g-n_g){S_g^2 \over n_g} V(Y^st)=g=1GNg(Ngng)ngSg2
均值估计的方差的估计 v ( y ˉ ) = ( 1 − f ) s 2 n v(\bar y) = (1-f){s^2 \over n} v(yˉ)=(1f)ns2 v ( y ˉ s t ) = ∑ g = 1 G W g 2 ( 1 − n g N g ) s g 2 n g v(\bar y_{st}) = \sum_{g=1}^GW_g^2(1-{n_g \over N_g}){s_g^2 \over n_g} v(yˉst)=g=1GWg2(1Ngng)ngsg2
总量估计的方差的估计 v ( N y ˉ ) = N ( N − n ) n s 2 v(N\bar y) = {N(N-n)\over n}s^2 v(Nyˉ)=nN(Nn)s2 v ( Y ^ s t ) = ∑ g = 1 G N g ( N g − n g ) s g 2 n g v(\hat Y_{st}) = \sum_{g=1}^GN_g(N_g-n_g){s_g^2 \over n_g} v(Y^st)=g=1GNg(Ngng)ngsg2
总体均值的置信区间的估计 [ y ˉ − u α 2 1 − f n s , y ˉ + u α 2 1 − f n s ] [\bar y - u_{\alpha \over 2}\sqrt {1-f \over n}s, \bar y + u_{\alpha \over 2}\sqrt {1-f \over n}s] [yˉu2αn1f s,yˉ+u2αn1f s]
总体总量的置信区间的估计 [ N y ˉ − u α 2 N ( N − n ) n s , N y ˉ + u α 2 N ( N − n ) n s ] [N\bar y - u_{\alpha \over 2}\sqrt {N(N-n) \over n}s, N\bar y + u_{\alpha \over 2}\sqrt {N(N-n) \over n}s] [Nyˉu2αnN(Nn) s,Nyˉ+u2αnN(Nn) s]
总体比例 P = A N P = {A \over N} P=NA P s t = ∑ g = 1 G W g P g P_{st} = \sum_{g=1}^GW_gP_g Pst=g=1GWgPg
具有所考虑特征的单元数 A A A A s t = ∑ g = 1 G N g P g A_{st} = \sum_{g=1}^GN_gP_g Ast=g=1GNgPg
总体方差 S 2 = N N − 1 P Q S^2 = {N\over N-1}PQ S2=N1NPQ
总体比例的估计 P ^ = p = a n \hat P = p = {a\over n} P^=p=na p s t = ∑ g = 1 G W g p g p_{st} = \sum_{g=1}^GW_gp_g pst=g=1GWgpg
具有特征的单元数的估计 A ^ = N p \hat A = Np A^=Np A ^ s t = ∑ g = 1 G N g p g \hat A_{st} = \sum_{g=1}^GN_gp_g A^st=g=1GNgpg
总体方差的估计 s 2 = n n − 1 p q s^2 = {n \over n-1}pq s2=n1npq
总体比例的估计的期望 E ( p ) = P E(p) = P E(p)=P E ( p s t ) = P s t E(p_{st}) = P_{st} E(pst)=Pst
总体比例的估计的方差 V ( p ) = P Q n N − n N − 1 V(p) = {PQ \over n}{N-n \over N-1} V(p)=nPQN1Nn V ( p s t ) = ∑ g = 1 G W g 2 V ( p g ) , V ( p g ) = ( 1 − f g ) S g 2 n g , S g 2 = N g N g − 1 P g Q g V(p_{st}) = \sum_{g=1}^GW_g^2V(p_g), V(p_g) = (1-f_g){S_g^2 \over n_g}, S_g^2 = {N_g \over N_g-1}P_gQ_g V(pst)=g=1GWg2V(pg),V(pg)=(1fg)ngSg2,Sg2=Ng1NgPgQg
单元数的估计的期望 E ( N p ) = A E(Np) = A E(Np)=A E ( A ^ s t ) = A s t E(\hat A_{st}) = A_{st} E(A^st)=Ast
单元数的估计的方差 V ( N p ) = N 2 P Q n N − n N − 1 V(Np) = {N^2PQ\over n}{N-n \over N-1} V(Np)=nN2PQN1Nn V ( A ^ s t ) = ∑ g = 1 G N g 2 V ( p g ) V(\hat A_{st}) = \sum_{g=1}^GN_g^2V(p_g) V(A^st)=g=1GNg2V(pg)
总体比例的估计的方差的估计 v ( p ) = N − n N ( n − 1 ) p q = 1 − f n − 1 p q v(p) = {N-n \over N(n-1)}pq = {1-f\over n-1}pq v(p)=N(n1)Nnpq=n11fpq v ( p s t ) = ∑ g = 1 G W g 2 v ( p g ) , v ( p g ) = ( 1 − f g ) s g 2 n g , s g 2 = n g n g − 1 p g q g v(p_{st}) = \sum_{g=1}^GW_g^2v(p_g), v(p_g) = (1-f_g){s_g^2 \over n_g}, s_g^2 = {n_g\over n_g-1}p_gq_g v(pst)=g=1GWg2v(pg),v(pg)=(1fg)ngsg2,sg2=ng1ngpgqg
单元数的估计的方差的估计 v ( N p ) = N ( N − n ) n − 1 p q v(Np) = {N(N-n) \over n-1}pq v(Np)=n1N(Nn)pq v ( A ^ s t ) = ∑ g = 1 G N g 2 v ( p g ) v(\hat A_{st}) = \sum_{g=1}^GN_g^2v(p_g) v(A^st)=g=1GNg2v(pg)
总体比例的置信区间 [ p − u α 2 1 − f n − 1 p q , p + u α 2 1 − f n − 1 p q ] [p - u_{\alpha \over 2}\sqrt {{1-f\over n-1}pq}, p + u_{\alpha \over 2}\sqrt {{1-f\over n-1}pq}] [pu2αn11fpq ,p+u2αn11fpq ]
单元数的置信区间 [ N p − u α 2 N ( N − n ) n − 1 p q , N p + u α 2 N ( N − n ) n − 1 p q ] [Np - u_{\alpha \over 2}\sqrt {{N(N-n) \over n-1}pq}, Np + u_{\alpha \over 2}\sqrt {{N(N-n) \over n-1}pq}] [Npu2αn1N(Nn)pq ,Np+u2αn1N(Nn)pq ]

注意

  • W g = N g N ≠ n g n W_g = {N_g \over N} \neq {n_g \over n} Wg=NNg=nng
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值