# NavList:

## A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding

### Compose Your Message

Message:αβγ
Message:abc
 Add Images & Files Posting Code: Name: Email:
Re: Linear regression and other tools
Date: 2018 Oct 12, 11:35 -0400
Hello Tony

You wrote:  I'm going to interleave the sights

In a 20 minute round of observations the travel of the vessel will likely be within the noise of CN.   Let us assume a velocity of 6 knots.  In 20 minutes, that means you traveled just 2 nautical miles.  The uncertainty of the horizon will yield errors on that order of magnitude.

Tony, you may find this older discussion useful.
In it, we discuss CN, accuracy and the clocked hat, but it is very germaine to the current discussion.  Which is, where exactly is your fix.  (As a very careful note: Frank took issue with my estimate of 10NM, and stated that CN, with care, can provide 3NM.)  CN is meant for ocean navigation.  When you get close to shore, switch to coastal piloting.

Squeezing a random error (not systemic, as you have noted!) isn't going to do much to improve your ocean navigation.  You are still within that same bounded box of uncertainty.

Let me flip it.  You managed to improve your accuracy, mid ocean, by 0.5NM.  Does it matter?

On Fri, Oct 12, 2018, 9:02 AM Tony Oz <NoReply_TonyOz@fer3.com> wrote:

Dear Antoine,

I am not going to extend anything beyond the measured interval, I'm going to interleave the sights (HCij) of each body: take HC11, then HC21, then HC31, then HC12, and so on, if there are three (i=3) bodies to sight. The j-index is the sight number within each body's sequence.

The value of the variable t should always be chosen within the interval between the first sight of the last body and the last sight of the first body, so that there will be NO extrapolations ever.

Warm regards,

Tony

60°N 30°E

PS

I see no point in taking several sights of the same body and only then to turn for the next one. My feeling is that unless I have to move the index arm significantly - there will be no chance the next sight will have an error of different sign, the error trend will stay along the series. Not enough randomness of error(s) is introduced between the measurements.

View and reply to this message

Browse Files

Drop Files

### Join NavList

 Name: (please, no nicknames or handles) Email:
 Do you want to receive all group messages by email? Yes No
You can also join by posting. Your first on-topic post automatically makes you a member.

### Posting Code

Enter the email address associated with your NavList messages. Your posting code will be emailed to you immediately.
 Email:

### Email Settings

 Posting Code:

### Custom Index

 Subject: Author: Start date: (yyyymm dd) End date: (yyyymm dd)