81 
First we have 
i=n 
ait 1 
DH, ,— X dH; , + dh, (n—1,n) 
i—1 
For the model p — 1,p we have the error 
i=p 
, b € 
DH, y) — - dH;_; + dh, 1, (330) 
im 
The residual error after correction in the control points is 
i=p p 
) , Ass x 7 4 Y L4 
RH, , Ly) T dH; =] + dh, (p—L.py n 2 dH, , + d "n (n—1,n) 
i i=1 
Hence 
yir A — pi p p i=n ; 
de nie — YY M à 
RH, Ln By - dH, 1 = dH; , +d D—1.0) ^^ 
i=l n i=p+1 
D x! . 
SE dh, (, 1,2) (331) 
n 
The weight number is 
; (n — pf p p* (n — p) m p 
Qry ry = n? ‘Qun gre n° ‘Qun T n° Qui 
or 
(n—p)p n® + PD 
fin = . T 332 
ry Ry a Quan a (332) 
h? h? 
From (257) we have Qgg — d and from (252) Q,,, T7 
y 2 p 
Hence 
: (n — p) ph? (à + pf e oi 
QOL ou ha + AT 33: 
UR ug nb? 2 n° b? 
This expression refers to the standard error of unit weight of the 
y-parallax measurements. 
6 
  
  
  
  
  
E E E 
I