Go ![]() | New ![]() | Find ![]() | Notify ![]() | Tools ![]() | Reply ![]() | |
His Royal Hiney![]() |
Seriously, do what I told you to do in Excel. AI is great but AI speaks to you in a way it thinks your level of understanding is. To use AI effectively or accurately, you need to be able to ask it the right questions. "It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946. | |||
|
Baroque Bloke![]() |
This is the question that I posed to ChatGPT: “What is the name of this function: (new - old) / |old|” What do you propose as “the right question”? Eagerly awaiting your response… Serious about crackers. | |||
|
His Royal Hiney![]() |
I wasn’t able to amend my post to add “or you’re not understanding the math” before you replied. Here’s another way I can show you that you’re misunderstanding the math or the mathematical relationships. 1) you agree that the following pair of numbers are one and the same right? 0.10 = 10%, 0.20 = 20%, 0.30 = 30%, etc. 2) we know they’re the same numbers because when you multiply any number by either one of the pair, you’re suppose to get the same answer. 3) the difference is that one of the pair is expressed as a “straight” number and the other is expressed as a percentage because by human convention, it’s easy for us to think in percentages or how many units of 1. Fifty per cent is 0.50, ten per cent is 0.10. That you’re supposed to get the same answer when multiplying by 0.25 or 25% tells you they’re one and the same. And you can get the same number displayed in either format in Excel as I described. "It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946. | |||
|
His Royal Hiney![]() |
I gave it to you in my very first answer on this thread. Relative Change or Relative Delta. The question I was referring to is the question you asked that gave you the wrong response to “percent something.” You do know it’s possible to reverse engineer a question to AI so that it gives you the answer you want, right? So it’s not about the “right” question but understanding both the question you are asking it and the answer it gives you. "It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946. | |||
|
Baroque Bloke![]() |
“Relative Change or Relative Delta.” is a statement, not a question. Still eagerly awaiting to hear what you say is “the right question” to pose to ChatGPT that’s better than: “What is the name of this function: (new - old) / |old|” Serious about crackers. | |||
|
Baroque Bloke![]() |
It is. Signing off this discussion. Serious about crackers. | |||
|
His Royal Hiney![]() |
This is the response I was referring to. Whatever question you asked ChatGPT to get that answer? That was the wrong question. The answer I gave you was what ChatGPT gave me with “Relative Delta” coming from me from my experience. "It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946. | |||
|
| Member |
Holy shit guys, you're going off the rails on something too trivial to argue about. The reason it is important to specify the scale used to represent the number is to avoid ambiguity. The same reason when we specify a temperature, we give the scale, Fahrenheit or Celsius, or other scale. If a convention is understood amongst the people using the number, the scale can be omitted, much like the local weatherman probably just says "It's 85 out today," because we conventionally use the Fahrenheit scale. And yes, a multiplication by a factor is required to convert a simple ratio into a percent scale, and this is precisely what Excel or any other spreadsheet is doing internally when it displays a number as a percent. This space intentionally left blank. | |||
|
| Member |
I'm so glad that I'm at a point in my life where none of this matters ... | |||
|
A Grateful American![]() |
"the meaning of life, is to give life meaning" ✡ Ani Yehudi אני יהודי Le'olam lo shuv לעולם לא עוד | |||
|
His Royal Hiney![]() |
Pipe Smoker, it wasn’t my intention to offend you; there was nothing personal or emotional in my responses to you. I simply saw a request for assistance from a forum member and, given my career of understanding numbers and numerical relationships, I gave my input. You asked for an appropriate question to ask AI. Below is what I crafted. You can see it is neutral in asking the question and it foresees the intended use: displaying the result for people to see — because if it’s just embedded somewhere in a calculation, not to be seen by anyone, then it doesn’t matter what you call it. If you’re going to call it anything, then it’s an easy assumption that the expectation is that people will see the number and refer to it. Also note that I asked how the number should be displayed because nothing in the formula is being changed based on how the result is displayed; it’s even as easy to display the result in a date format in Excel — not that it would make human sense. Lastly, you can do as I did and asked AI. I fed the same prompt to Grok, Gemini, and ChatGPT and compare your results to mine. As I said, it’s in a large part in how you ask the question. Prompt: “When displaying the result of the following formula “(New - Old) / ABS ( Old)” for optimal human viewing, should it be displayed in its basic numerical format with the appropriate number of trailing decimal points or as a percentage, also with the appropriate number of trailing decimal points? Why or why not? What is the standard practice? Recommend appropriate names for the result.” You can also add this additional question in the end which I did not do: “ If the result is displayed as a percentage, is the formula changed?” Grok: The formula (New - Old) / ABS(Old) calculates a relative change or difference, often used to express how much a value has shifted proportionally to the original (old) value, while ensuring the denominator is positive to handle cases where the old value might be negative (e.g., in financial losses or temperature scales). For optimal human viewing, it should be displayed as a percentage (e.g., “15.25%”) rather than in basic numerical format (e.g., “0.1525”). Here’s the reasoning: • Why percentage? Humans generally find percentages more intuitive and relatable than decimal fractions, especially for comparative or proportional metrics. A percentage format allows quick mental interpretation—like recognizing a “20% increase” as substantial without needing to multiply by 100 mentally—reducing cognitive load and errors in comprehension. It aligns with everyday language (e.g., “sales grew by 10%”) and is visually cleaner in reports, charts, or dashboards. Trailing decimal points should be included based on precision needs (e.g., 1-2 decimals for general audiences, more for scientific contexts), but always with the “%” symbol and a sign if negative (e.g., “-5.0%”). • Why not basic numerical? Raw decimals (e.g., 0.05 or -0.1234) are less immediately meaningful, requiring extra mental steps to contextualize as a proportion. They’re better suited for machine processing or intermediate calculations, not end-user display. • Standard practice: In fields like finance, economics, statistics, business analytics, and data visualization (e.g., per guidelines from sources like the APA Style Manual, Excel defaults, or Tableau best practices), such relative change metrics are routinely formatted as percentages. This is evident in stock market reports (e.g., “stock rose 2.3%”), economic indicators (e.g., GDP growth), or scientific papers discussing rates of change. Tools like spreadsheets (Google Sheets, Excel) often auto-format these as percentages when labeled as such, and data viz libraries (e.g., Matplotlib, ggplot) encourage it for readability. Exceptions might occur in pure mathematical or programming contexts where precision without symbols is preferred, but for “human viewing,” percentage is the norm. Recommended names for the result (choose based on context for clarity): • “Percentage Change” (most common and straightforward, implying the proportional shift). • “Relative Change” (if emphasizing the fractional nature, sometimes used interchangeably). • “Growth Rate” (if the change is positive or in a growth-oriented context like sales or population). • “Percent Difference” (useful when comparing two similar values, like benchmark vs. actual). • “Absolute Relative Change” (to highlight the use of ABS(Old) for handling negative old values). Gemini: You should display the result of the formula (New - Old) / |Old| as a percentage for optimal human viewing. Why a Percentage? Displaying the result as a percentage is the standard practice because it intuitively represents a rate of change or relative difference. A percentage provides a context for the magnitude of the change, making it much easier for people to understand and compare. For example, a change from 100 to 120 is an increase of 20. The formula gives you 0.20, but most people can quickly recognize that a 20% increase is a significant gain. Conversely, a basic numerical format like 0.05 or 0.25 lacks immediate context and requires an extra mental step for the viewer to interpret its meaning. A value like 0.005 could represent a change of 0.5% or a change of 0.5 depending on the context. By displaying it as a percentage, you eliminate this ambiguity. Naming Conventions The result of this calculation is commonly referred to as Percent Change or Percentage Change. Other appropriate names include Rate of Change, Growth Rate, or Relative Difference. When the result is a positive number, it can be called Percent Increase; when it is negative, it's called Percent Decrease. ChatGPT: When it comes to displaying the result of the formula (New - Old) / ABS(Old), it's generally better to present it as a percentage rather than a pure numerical format. Here’s why, along with some recommendations: ### Why Use a Percentage? 1. **Intuitive Understanding**: Percentages are intuitive for most people. They provide a clear sense of proportion and scale relative to the original value. For instance, saying "a 20% increase" is more universally understood than saying "0.2". 2. **Comparability**: Percentages make it easier to compare changes across different scenarios or datasets. This is especially useful in reports, presentations, or discussions where stakeholders might need to quickly grasp context and make comparisons. 3. **Standard Practice**: In most analytical contexts—financial analysis, sales reports, scientific data—the change is often discussed in percentage terms. This practice helps maintain consistency and aligns with what audiences typically expect. ### Decimal Points - The number of trailing decimal points should align with the precision needed for your audience or the sensitivity of the data. For general purposes, one or two decimal places are usually sufficient when expressing percentages (e.g., 15.35%). ### Appropriate Names for the Result Here are a few name suggestions for the result of your formula: 1. **Percentage Change**: Simple and straightforward; it communicates exactly what the figure represents. 2. **Rate of Change**: A bit more technical, but still widely understood. 3. **Relative Change**: This name emphasizes the change relative to the original value. 4. **Percent Variation**: Highlights the concept of variance from the old value. ### Conclusion Displaying the result as a percentage is the standard and preferred approach for optimal human understanding. It combines clarity, comparability, and consistency, which are key factors in effective data communication. "It did not really matter what we expected from life, but rather what life expected from us. We needed to stop asking about the meaning of life, and instead to think of ourselves as those who were being questioned by life – daily and hourly. Our answer must consist not in talk and meditation, but in right action and in right conduct. Life ultimately means taking the responsibility to find the right answer to its problems and to fulfill the tasks which it constantly sets for each individual." Viktor Frankl, Man's Search for Meaning, 1946. | |||
|
Baroque Bloke![]() |
Thanks for your thoughtful and courteous post Rey. However, for my purposes, I still like ChatGPT’s reply to this simple, unambiguous query: “What is the name of this function? (new - old) / |old|” “This measure is the signed relative change (commonly called the relative change or fractional change); when multiplied by 100 it is the percent change. The absolute value in the denominator ensures the change is scaled by the magnitude of the original value regardless of its sign.” I consider that reply to be concise, comprehensive, and, IMO, wholly correct. It well describes my rchng (relative change) function, which I’ve used since I created it in February, 2023. And it’s also in general agreement with my name choice. And it doesn’t chide me for creating an “relative change” function rather than a “percent change” function. It simply provides the name of the function that I DID create, which is the response I wanted. I’ve long had another function, “percentify(n, d)”, that can utilize my “rchng(old, new)” function to return a percent change rounded to “d” decimal places. And adorned with a trailing ‘%’ character. Edited to add two examples of my percentify function used with my rchng function: percentify(rchng(1000, 1001), 2) Returns +0.10% percentify(rchng(1001, 1000), 2) Returns -0.10%This message has been edited. Last edited by: Pipe Smoker, Serious about crackers. | |||
|
| Powered by Social Strata | Page 1 2 |
| Please Wait. Your request is being processed... |
|

