Proving Natural Log Proof: ln|1+σx|

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
10 replies · 2K views
Nusc
Messages
752
Reaction score
2

Homework Statement


Prove the following statement:
[tex]ln|1+\sigma x | = \frac{1}{2} ln|1-x^2| + \frac{\sigma}{2} ln| \frac{ |1+x|}{|1-x|}[/tex]

Homework Equations

The Attempt at a Solution


Starting from right to left would be easier:
[tex] = \frac{1}{2} ln|(1+x)(1-x)| + \frac{\sigma}{2} ln| 1+x| - \frac{\sigma}{2} ln(1-x) \\<br /> = \frac{1}{2} [ ln(1+x) + ln(1-x) + \sigma ln(1+x) - \sigma ln(1-x)] \\<br /> =\frac{1}{2} [ln|(1+x)(1+\sigma)| + ln|(1-x)(1-\sigma)|] \\<br /> = \frac{1}{2} [ ln |(1+x)^{1+\sigma} (1-x)^{1-\sigma} |]\\<br /> = \frac{1}{2} ln |( 1+x ) (1+x)^\sigma \frac{1+x}{(1-x)^\sigma}|][/tex]
Then I get stuck. Does anyone know what I'm missing?
 
Physics news on Phys.org
Do you know of any identities that would give me the left hand side? Any references? I cannot find anything online. Calculus textbooks just give integrals and not identities for logs.

I need to take the natural log of the following expression.
[tex]\frac{1}{2}( C + D ) + \frac{1}{2}( C - D ) \sigma[/tex]

The idea was to use that identity, then rearranging above
[tex]\frac{1}{2}( C + D ) [ 1 + \frac{(C-D)}{(C+D)} \sigma ][/tex]

The answer is

[tex] ln2 + \frac{1}{2}ln|4CD|+\frac{\sigma }{2} ln|\frac{C}{D} |[/tex]

Without that identity, It's pretty hard to achieve that result. Do you know of anything I can use?

What's really important is that I at least get this form: [tex]\frac{\sigma }{2} ln|\frac{C}{D} |[/tex]
 
Last edited:
Nusc said:

Homework Statement


Prove the following statement:
[tex]ln|1+\sigma x | = \frac{1}{2} ln|1-x^2| + \frac{\sigma}{2} ln| \frac{ |1+x|}{|1-x|}[/tex]

Homework Equations

The Attempt at a Solution


Starting from right to left would be easier:
[tex] = \frac{1}{2} ln|(1+x)(1-x)| + \frac{\sigma}{2} ln| 1+x| - \frac{\sigma}{2} ln(1-x) \\<br /> = \frac{1}{2} [ ln(1+x) + ln(1-x) + \sigma ln(1+x) - \sigma ln(1-x)] \\<br /> =\frac{1}{2} [ln|(1+x)(1+\sigma)| + ln|(1-x)(1-\sigma)|] \\<br /> = \frac{1}{2} [ ln |(1+x)^{1+\sigma} (1-x)^{1-\sigma} |]\\<br /> = \frac{1}{2} ln |( 1+x ) (1+x)^\sigma \frac{1+x}{(1-x)^\sigma}|][/tex]
Then I get stuck. Does anyone know what I'm missing?
You appear to be a step or two away.

Added in Edit:
Never mind! I was looking at the wrong side of the equation.
 
Last edited:
Nusc said:

Homework Statement


Prove the following statement:
[tex]ln|1+\sigma x | = \frac{1}{2} ln|1-x^2| + \frac{\sigma}{2} ln| \frac{ |1+x|}{|1-x|}[/tex]

Homework Equations

The Attempt at a Solution


Starting from right to left would be easier:
[tex] = \frac{1}{2} ln|(1+x)(1-x)| + \frac{\sigma}{2} ln| 1+x| - \frac{\sigma}{2} ln(1-x) \\<br /> = \frac{1}{2} [ ln(1+x) + ln(1-x) + \sigma ln(1+x) - \sigma ln(1-x)] \\<br /> =\frac{1}{2} [ln|(1+x)(1+\sigma)| + ln|(1-x)(1-\sigma)|] \\<br /> = \frac{1}{2} [ ln |(1+x)^{1+\sigma} (1-x)^{1-\sigma} |]\\<br /> = \frac{1}{2} ln |( 1+x ) (1+x)^\sigma \frac{1+x}{(1-x)^\sigma}|][/tex]
Then I get stuck. Does anyone know what I'm missing?

Your desired result is false. For small ##|x|## we can expand both sides in a series in ##x##. For ease of writing, I will use the symbol ##a## instead of ##\sigma##. Let
$$L(x) = \ln (1+ ax) , \;\; R(x) = \frac{1}{2} \ln(1-x^2) + \frac{a}{2} \ln \left(\frac{1+x}{1-x}\right).$$
We have
$$L(x) = ax - \frac{1}{2} a^2 x^2 + \frac{1}{3} a^3 x^3 - \frac{1}{4} a^4 x^4 + O(x^5),$$
but
$$R(x) = ax - \frac{1}{2} x^2 + \frac{1}{3} a^3 x^3 - \frac{1}{4} x^4 + O(x^5).$$
Note that in ##L(x), R(x)## I have omitted the absolute-value signs; those signs are redundant when ##|x| < 1##, which is true for small ##|x|##.
 
Was there a typo? Is there a way to determine if there is one?

Or is there an easier way to simplify the natural log of the expression?

[tex]\frac{1}{2}( C + D ) + \frac{1}{2}( C - D ) \sigma[/tex]

with obtaining
[tex]\frac{\sigma }{2} ln|\frac{C}{D} |[/tex] at least in the final answer?[tex] ln [ \frac{1}{2} D ( \frac{C}{D} +1 ) + \frac{\sigma}{2} D ( \frac{C}{D} -1) ][/tex]
 
Nusc said:
Was there a typo?
For it even to be approximately true you need to change the 1/2 factor in the first expression on the right to σ/2. That makes it more symmetric, so a lot more reasonable.
It still is not going to be generally true. The roles of the σ on the left and the σ's on the right are too different. I would guess you are only supposed to show this as asymptotic behaviour for small x. (It is equivalent to (1+x)n approximating enx.)
 
Is there a table of asymptotic expansions or reference I can refer to to get the desired result? I'm kind of stumped.
 
Nusc said:
Is there a table of asymptotic expansions or reference I can refer to to get the desired result? I'm kind of stumped.
First step is to correct the RHS of the statement to be proved. See the first para of post #8.
Next, as Ray points out, on the assumption that |x|<1 and |σx|<1, all those modulus signs can be thrown away.
Then you can exponentiate both sides to get rid of the logarithms. Post what you get.