Trying to find percent change in revenue from percent change in profits and percent change in cost.

84 Views Asked by At

I'm trying to find the percent change in revenue, $I$ from the percent change in profits $P$ and the percent change in costs $E$.

Suppose we are given that $R_1, R_2, E_1$ and $E_2$ are positive.

We have three equations:

$$ \begin{aligned} (1)&\quad P = \frac{(R_2 - E_2) - (R_1 - E_1)}{R_2-E_2}, \quad 0 < P < 1 \\ (2)&\quad C = \frac{E_2 - E_1}{E_2}, \quad 0 < C < 1 \\ (3)&\quad I = \frac{R_2 - R_1}{R_2}, \quad 0 < I < 1 \\ \end{aligned} $$

We know the values of P and C, but not I.

Also assume we know the values of $R_1-E_1 = M_1$ and $R_2-E_2 = M_2$, but we do not know anything about the absolute change in costs or revenue.

Is there a way to find the value of $I$ from the values of $P$ and $C$?

If not, is there a way to approximate the value of $I$?

1

There are 1 best solutions below

0
On

I think the approximation is complicated, I did the following breakdown:

There are 2 cases, derived from your comment that a lower bound of zero is not acceptable, but first of all let's write $I$ as:
$$I = 1 - \frac{R_1}{R_2}$$

Case 1:
$I$ is arbitrarily close to zero. Yes, it can be, cold shower right at the beginning. For this, we have $R_1 < R_2$ in such a way, that $R_1 \approx R_2 \Rightarrow \frac{R_1}{R_2} \approx 1$. As $R$'s are upper bounds for $E$'s, we have further 2 cases, but we may observe that this bound doesn't matter:
Case 1A:
$E_1 < E_2$ in such a way, that $E_1 \approx E_2 \Rightarrow \frac{E_1}{E_2} \approx 1$. This means that $C$ is arbitrarily close to zero by the same logic for $I$. This relation between $E_1$ and $E_2$ is possible independently of $R_1$ and $R_2$. And implies $M_1 < M_2$ in such a way, that $M_1 \approx M_2$ - arbitrarily close.
Case 1B:
$E_1 << E_2$. This means that $C$ is arbitrarily close to $1$. This relation between $E_1$ and $E_2$ is possible independently of $R_1$ and $R_2$. This implies $M_1 >> M_2$ - arbitrarily far, however this is not possible, as $P > 0 \Rightarrow M_1 < M_2$.
Case summary: having $M_1 < M_2$ and $C$ sufficiently close to $1$ should give a lower bound > 0, otherwise the lower bound is 0.

Case 2:
$I$ is arbitrarily close to $1$. For this, we have $R_1 << R_2$. As $R$'s are upper bounds for $E$'s, we have further 2 cases, but we may observe that this bound doesn't matter:
Case 2A:
$E_1 < E_2$ in such a way, that $E_1 \approx E_2 \Rightarrow \frac{E_1}{E_2} \approx 1$. This means that $C$ is arbitrarily close to zero. This relation between $E_1$ and $E_2$ is possible independently of $R_1$ and $R_2$. And implies $M_1 << M_2$ - arbitrarily far.
Case 2B:
$E_1 << E_2$. This means that $C$ is arbitrarily close to $1$. This relation between $E_1$ and $E_2$ is possible independently of $R_1$ and $R_2$. This implies either $M_1 < M_2$ in such a way, that $M_1 \approx M_2 \Rightarrow \frac{M_1}{M_2} \approx 1$ - arbitrarily close, or $M_1 << M_2$ - arbitrarily far.
Case summary: having $M_1 < M_2$ and $C$ sufficiently close to zero should give an upper bound < 1, otherwise the upper bound is 1.

Total summary: possibly in a contradictory way, $C$ can give both a lower and an upper bound. This is because there is no one bounded $I$, but there are certain cases - so far I covered:
Case 1A: $C \approx 0$ and $P \approx 0 \Rightarrow I \approx 0$
Case 1B: $C \approx 1$ and $P$ any $\Rightarrow I > 0$ - lower bound can be found
Case 2A: $C \approx 0$ and $P \approx 1 \Rightarrow I \approx 1$
Case 2B: $C \approx 1$ and $P$ any $\Rightarrow I < 1$ - upper bound can be found

Cases 1B and 2B tell us, that with the same conditions, we can find bounds - this is how it goes:
$C \approx 1 \Rightarrow E_2 \approx E_1 \Rightarrow P = \frac{(R_2 - E_2) - (R_1 - E_1)}{(R_2 - E_2)} \approx 1 - \frac{(R_1 - E_1)}{(R_2 - E_2)}$ - this should be compared with the formula of $I = 1 - \frac{R_1}{R_2}$:
If $R_2 - E_2 = M_2 \approx R_2 \Rightarrow P > I$, that is, an upper bound.
If $R_1 - E_2 \approx R_1 - E_1 = M_1 \approx R_1 \Rightarrow P < I$, that is, a lower bound.
If both $M_2 \approx R_2$ and $M_1 \approx R_1 \Rightarrow P \approx I$, an approximation.
However, we will never know these, as $R$'s and $E$'s are not given, so I think it is impossible to practically know the bounds. At this point I finish the breakdown, still cases are not covered, e.g. $C \approx 0$ and $P$ any, or $P \approx 1$ and $C$ any, I leave it to you to finish (after correcting any mistakes I made with the above), but I guess practically none of those gave any bounds either.