Fascinating Properties: Excluding 998 from 0.001002003004...

  • Context: MHB 
  • Thread starter Thread starter Jameson
  • Start date Start date
  • Tags Tags
    Properties
Click For Summary

Discussion Overview

The discussion revolves around the properties of the fraction \(\frac{1}{998001}\) and its decimal representation, which lists every three-digit number except for 998. Participants explore the generalization of this property to other fractions and bases, examining the underlying characteristics and potential patterns in their decimal expansions.

Discussion Character

  • Exploratory
  • Technical explanation
  • Mathematical reasoning

Main Points Raised

  • Some participants note that \(\frac{1}{998001}\) has a decimal representation of 0.001002003004..., excluding the number 998.
  • One participant conjectures that the family of such fractions can be expressed as \(\frac{1}{(10^n-1)^2}\), where the decimal representation contains all \(n\)-digit numbers except for \(10^n-2\).
  • Another participant confirms that \(\frac{1}{9999^2}\) similarly gives all four-digit numbers, suggesting a consistent pattern across different denominators.
  • A participant proposes that this property may generalize to an arbitrary base \(b\), where \(\frac{1}{(b^n - 1)^2}\) contains all \(n\)-digit numbers in that base except for \(b^n - 2\).
  • Some participants express uncertainty about the generalization to other bases, noting that they have only checked a couple of bases.
  • One participant expresses interest in the topic without contributing additional technical details.

Areas of Agreement / Disagreement

Participants generally agree on the interesting properties of the fractions discussed, but multiple competing views remain regarding the generalization to other bases and the validity of the conjectures presented. The discussion remains unresolved on certain aspects, particularly the broader applicability of the identified patterns.

Contextual Notes

Some limitations include the dependence on specific definitions of the fractions and the conjectural nature of the generalizations proposed, which have not been thoroughly verified across all bases.

Jameson
Insights Author
Gold Member
MHB
Messages
4,533
Reaction score
13
[math]\frac{1}{998001}[/math] has some interesting properties. It will list out every 3 digit number except for 998.

It has the form 0.001002003004...

Here's a video on some more details if you're interested.

[video=youtube;daro6K6mym8]998001[/video]
 
Physics news on Phys.org
soroban once posted this observation and asked:

This is just one of a family of such fractions.
Can you determine the underlying characteristic?

I looked at the factorization of the denominator, and found:

$\displaystyle 998001=999^2$

So, next I looked at:

$\displaystyle\frac{012345679}{999999999}=\frac{1}{9^2}$

Thus, I conjecture that the family you speak of is:

$\displaystyle\frac{1}{(10^n-1)^2}$ where $\displaystyle n\in\mathbb N$

where the decimal representation contains all of the $n$ digit numbers except $\displaystyle 10^n-2$ and the period is $\displaystyle n(10^n-1)$.
 
I had a feeling when I was posting that soroban might have already said something about this :(

Yes, you are correct about this form. For example [math]\frac{1}{9999^2}[/math] gives all of the 4 digit numbers.

In the video I posted in the OP they discussed an easy way to write recurring decimals as fractions that I'm sure we are all familiar with this method but it's still worth writing for those who aren't.

If you want to write some repeating decimal all you do is write the string as the numerator to a fraction and then in the denominator write the same number of 9's. For example, if you want to write [math].\overline{12436298}[/math] as a fraction. It's simply [math]\frac{12436298}{99999999}[/math]. Of course this method can be derived quite easily. If [math]x=.12436298[/math] then $100000000x=12436298.\overline{12436298}$ so $9999999x=12436298$ and we can solve for x directly. However it's nice to know a shortcut to not have to calculate it "the long way" everytime.
 
Last edited:
This seems to generalize to an arbitrary base $b$, where:

$$\frac{1}{(b^n - 1)^2}$$

has representation in base $b$ containing all $n$-digit numbers (in base $b$) except $b^n - 2$ and with period $n(b^n - 1)$.

But I only checked a couple of bases, so this could be wrong.
 
Bacterius said:
This seems to generalize to an arbitrary base $b$, where:

$$\frac{1}{(b^n - 1)^2}$$

has representation in base $b$ containing all $n$-digit numbers (in base $b$) except $b^n - 2$ and with period $n(b^n - 1)$.

But I only checked a couple of bases, so this could be wrong.

I believe you are right, as the method Jameson outlined above can be generalized to any valid radix.
 
Not much to contribute to this conversation, I was just popping into say that this is really interesting!
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 37 ·
2
Replies
37
Views
4K
  • · Replies 3 ·
Replies
3
Views
1K
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 10 ·
Replies
10
Views
5K
  • · Replies 2 ·
Replies
2
Views
7K
  • · Replies 6 ·
Replies
6
Views
2K