Donate to Remove ads

Got a credit card? use our Credit Card & Finance Calculators

Thanks to johnstevens77,Bhoddhisatva,scotia,Anonymous,Cornytiv34, for Donating to support the site

Unilever - Half-year Report

For discussion of the practicalities of setting up and operating income-portfolios which follow the HYP Group Guidelines. READ Guidelines before posting
Forum rules
Tight HYP discussions only please - OT please discuss in strategies
Gengulphus
Lemon Quarter
Posts: 4255
Joined: November 4th, 2016, 1:17 am
Been thanked: 2628 times

Re: Unilever - Half-year Report

#430129

Postby Gengulphus » July 24th, 2021, 3:38 pm

Arborbridge wrote:Thanks for your comments and your acute observational powers which have definitely spotted revealed a problem. Whether this is a big can of worms or just a minor one, I'm not sure!

I cut and paste prices from Yahoo, and choose "weekly" on the historic data tab. In this case the dates ran 2nd,9th, 16th, 23rd - each a Monday - so they do actually miss the sharp low you mention, and the max yield was missed.
That accounts for some of the difference, but there's something else which is worrying me. When I revisit prices on Yahoo, for example, to update a chart, I've sometimes noticed they are not necessarily the same as they were, and so it is here. When I went back to check subsequent to your observation, I find the prices are not the same as the ones I cut and pasted on the previous occasion. It's a though revision takes place a while afterwards, but I have no knowledge of when this happens.

Here's a cut and past from my spreadsheet which gives the price as it was originally on Yahoo, then the price shown for the same dates today. I copy the closing price in every case. The next columns are the yield plotted (based on the previous four dividend total of 148.18p) and the corrected yield from the revised prices. If I update the chart now, the highest yield of 3.98% would be a smidgen above the January 21 yield - though it would still miss the best yield which occurred.

Image

I'm 99% certain I've spotted what's going on here, and I'm afraid it's probably going to result in you kicking yourself!

I checked the "now" column in your table against figures from the FT, and they matched its closing prices for Monday March 2nd, 9th, 16th, 23rd and 30th, 2020 (apart from having no decimal places and the FT prices having one decimal place). But the "then" figures bore no relationship I could see to the prices in March 2020 beyond being in roughly the same high-3000s-to-low-4000s range.

And then I noticed the years in your date column, and tried getting the FT's closing prices for March 2021, but the "then" column also didn't match those prices for the 2nd, 9th, 16th, 23rd and 30th... But I did spot a nearby closing price of 3891.0p, for Friday 5th March, 2021 - and after extending the date range a day into April, it turned out that your "then" column matches this year's prices for Friday March 5th, 12th, 19th, 26th and Thursday April 1st, 2021 (Friday April 2nd was the Good Friday bank holiday). So it looks to me as though you originally inadvertently picked up the wrong year's closing prices for the weeks containing March 2nd, 9th, 16th, 23rd and 30th! And that your "now" column is the closing prices for the days, not for the weeks containing the days.

Gengulphus

ADrunkenMarcus
Lemon Quarter
Posts: 1584
Joined: November 5th, 2016, 11:16 am
Has thanked: 672 times
Been thanked: 479 times

Re: Unilever - Half-year Report

#430181

Postby ADrunkenMarcus » July 24th, 2021, 9:29 pm

scrumpyjack wrote:This will be a test of the supposed strength of Unilever's brands. The whole point a a strong brand is that it gives the manufacturer pricing power, so if their brands are as strong as claimed they should be able to push up prices to recover these costs.


I think that's entirely true, however we'll really need to wait and see over a longer period. I am not sure, for example, to what extent they have fixed price agreements and how long they run for?

Best wishes


Mark.

Dod101
The full Lemon
Posts: 16629
Joined: October 10th, 2017, 11:33 am
Has thanked: 4343 times
Been thanked: 7534 times

Re: Unilever - Half-year Report

#430183

Postby Dod101 » July 24th, 2021, 9:37 pm

I frankly think that there is nothing to worry about. We are all micro analysing a business that has been driving forward in its current form for say 90 yeas or more. What is 6 months? We are worrying about market noise.

Dod

moorfield
Lemon Quarter
Posts: 3523
Joined: November 7th, 2016, 1:56 pm
Has thanked: 1546 times
Been thanked: 1402 times

Re: Unilever - Half-year Report

#430190

Postby moorfield » July 24th, 2021, 10:32 pm

Dod101 wrote:I frankly think that there is nothing to worry about. We are all micro analysing a business that has been driving forward in its current form for say 90 yeas or more. What is 6 months? We are worrying about market noise.

Dod


Well said. I think micro analysing sums up 90% of the traffic in this place ...

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430207

Postby Arborbridge » July 25th, 2021, 7:50 am

Gengulphus wrote:I'm 99% certain I've spotted what's going on here, and I'm afraid it's probably going to result in you kicking yourself!



Gengulphus


I certainly am kicking myself!: we'd better start again.

Yesterday, I inadvertently copied the March 2021 prices in place of the March 2020. I should have twigged because the dates were incorrect in the original and I altered them manually to make them fit the "mondays" - I just assumed the dates were wrong in the spreadsheet whereas in fact I was viewing the wrong year!

The table for March 2020 should look like this:-

Image

This gives the highest yield as 3.83% on the 16th March, and would make the two peaks almost the same. This still doesn't give me an explanation as to why the "then" and "now" prices are different, where "now" means the closing price I can see today on Yahoo, and "then" was what I copied and pasted at some time in the past from Yahoo.

Arb,

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430208

Postby Arborbridge » July 25th, 2021, 7:55 am

moorfield wrote:
Dod101 wrote:I frankly think that there is nothing to worry about. We are all micro analysing a business that has been driving forward in its current form for say 90 yeas or more. What is 6 months? We are worrying about market noise.

Dod


Well said. I think micro analysing sums up 90% of the traffic in this place ...



It is just possible it goes with the territory? - the type of person who tries to understand what's going on with share prices will also be likely to want to probe further and further? Perhaps we just enjoy it, though it's nice when the pain stops. :lol:

Naturally, not everyone is in the obsessively analytical camp and some float in and out of it. I suspect those who came from an accounting background want to solve a problem to the nearest penny, and those from a scientific background have the urge to tease apart any problem until the understand it.

Arb,

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430213

Postby Arborbridge » July 25th, 2021, 8:48 am

Out of curiosity, I decided to find out whether Yahoo prices do change, or whether I'm imagining it - this is in the context of my conversation with Gengulphus about the error in my yield chart.

Below is a column of closing prices pasted from Yahoo on the day that I revised my chart this week, marked "Then" with the next column being marked "Today". Today being early this morning.
The previous past and copy would have been March 2021, so prices from January to March would have been copied at that time: price after March would have been copied this week and added to the column.

The final column on the right shows any changes, and indeed there are two.


I'm not clear what is going on, or whether I am making some of error myself, but there does seem to be an occasional problem.

Arb.

moorfield
Lemon Quarter
Posts: 3523
Joined: November 7th, 2016, 1:56 pm
Has thanked: 1546 times
Been thanked: 1402 times

Re: Unilever - Half-year Report

#430295

Postby moorfield » July 25th, 2021, 4:17 pm

Arborbridge wrote:
moorfield wrote:
Dod101 wrote:I frankly think that there is nothing to worry about. We are all micro analysing a business that has been driving forward in its current form for say 90 yeas or more. What is 6 months? We are worrying about market noise.

Dod


Well said. I think micro analysing sums up 90% of the traffic in this place ...



It is just possible it goes with the territory? - the type of person who tries to understand what's going on with share prices will also be likely to want to probe further and further? Perhaps we just enjoy it, though it's nice when the pain stops. :lol:

Naturally, not everyone is in the obsessively analytical camp and some float in and out of it. I suspect those who came from an accounting background want to solve a problem to the nearest penny, and those from a scientific background have the urge to tease apart any problem until the understand it.



I guess you're right. It's always struck me as strange that HYP-P attracts the most traffic here, the antithesis of what it should be doing (buy shares, leave them alone, and live your life doing/posting about other stuff). But lest I kick off yet another discussion about what would replace it (my impression is that this place pretty much replicates what I've always thought a home counties village share club would look and sound like), I'll stop right there!

Gengulphus
Lemon Quarter
Posts: 4255
Joined: November 4th, 2016, 1:17 am
Been thanked: 2628 times

Re: Unilever - Half-year Report

#430335

Postby Gengulphus » July 25th, 2021, 9:19 pm

Arborbridge wrote:Yesterday, I inadvertently copied the March 2021 prices in place of the March 2020. I should have twigged because the dates were incorrect in the original and I altered them manually to make them fit the "mondays" - I just assumed the dates were wrong in the spreadsheet whereas in fact I was viewing the wrong year!

The table for March 2020 should look like this:-

Image

This gives the highest yield as 3.83% on the 16th March, and would make the two peaks almost the same. This still doesn't give me an explanation as to why the "then" and "now" prices are different, where "now" means the closing price I can see today on Yahoo, and "then" was what I copied and pasted at some time in the past from Yahoo.

Checking up using the FT again, your "now" prices are the closing prices for the days Monday March 2nd, 9th, 16th, 23rd and 30th, 2020, give or take the odd 0.5p. Your "then" prices are the closing prices for the Monday-to-Friday weeks containing those days - which are the closing prices for the last trading days in those weeks. The Easter bank holiday was some way into April in 2020, and so those final trading days were all Fridays: March 6th, 13th, 20th, 27th and April 3rd, 2020. With share prices having been particularly volatile around then, just one trading day could make a considerable difference to prices, and so there doesn't look to be any particular relationship between the columns...

Whether the difference is due to you doing something different between then and now, or to Yahoo changing their weekly prices facility over from giving closing prices for Monday-to-Friday weeks to closing prices for days at weekly intervals, I cannot tell...

Arborbridge wrote:Out of curiosity, I decided to find out whether Yahoo prices do change, or whether I'm imagining it - this is in the context of my conversation with Gengulphus about the error in my yield chart.

Below is a column of closing prices pasted from Yahoo on the day that I revised my chart this week, marked "Then" with the next column being marked "Today". Today being early this morning.
The previous past and copy would have been March 2021, so prices from January to March would have been copied at that time: price after March would have been copied this week and added to the column.

The final column on the right shows any changes, and indeed there are two.


I'm not clear what is going on, or whether I am making some of error myself, but there does seem to be an occasional problem.

Some more looking at the FT prices says that

* 3733 was the closing price on Friday, February 26th, 2021.
* 3920 was the closing price on Tuesday, February 23rd, 2021.
* 4144 was the closing price on Friday, July 23rd, 2021.
* 4050.5 was the closing price on Thursday, July 22nd, 2021.

Hypothesis: Yahoo gives you the closing price for the Monday-to-Friday week containing the date you specify - normally a Friday, but bank holidays can push it earlier - and so can the week not yet being complete.

If that hypothesis is right, your table could be explained by you having updated your data at some time between the market closes on Tuesday February 23rd and Wednesday February 24th earlier this year, then updated it again at some time between the market closes on Thursday July 22nd and Friday July 23rd this last week (only doing the dates that you didn't already have data for, of course), and then done the "now" check today for the full range of dates. You've already said you did the last part today, so that's confirmed. The second part is more specific than your "the day that I revised my chart this week", but seems entirely consistent with you having posted the chart at 07:16 on Friday July 23rd. Does the first part seem reasonably consistent with your recollection of when you previously updated your data for the chart? (I'm obviously not expecting you to remember exact dates 5 months ago! - but does about 5 months ago seem right?)

You could also try to get additional confirmation by asking Yahoo for the closing price for Monday July 26th (tomorrow) after the market close tomorrow and further days this week - if the hypothesis is right, it's likely to change on each of Tuesday, Wednesday, Thursday and Friday, and then stop changing. (But I'm rather doubtful this additional confirmation is necessary - it may well count as gilding the lily!)

Finally, as regards what to do about it assuming this is what is going on, I can see various alternative possibilities (no need to do more than one of them):

* Switch over to using daily data - this will be less likely to miss seeing sharp yield spikes (*).

* Switch over to using weekly data for Fridays rather than Mondays.

* Only update your data at weekends.

* Each time you update your data, mark the last data point in some way - I would probably give it a yellow background colour in the spreadsheet, as a reminder that it needs checking/updating. Update the data point that was left marked by the last update, remove the mark from it, then add further data points as far as you can and mark the last data point added.

(*) Though for the best way of catching sharp yield upward spikes, use "low" share prices instead of "closing" share prices, and for the best way of catching sharp yield downward spikes, use "high" prices. Those will guarantee catching the even very narrow yield spikes using weekly or even lower-frequency data - though it's a bit awkward if you want to show both types of spike...

Gengulphus

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430393

Postby Arborbridge » July 26th, 2021, 8:04 am

Gengulphus wrote:Hypothesis: Yahoo gives you the closing price for the Monday-to-Friday week containing the date you specify - normally a Friday, but bank holidays can push it earlier - and so can the week not yet being complete.

If that hypothesis is right, your table could be explained by you having updated your data at some time between the market closes on Tuesday February 23rd and Wednesday February 24th earlier this year, then updated it again at some time between the market closes on Thursday July 22nd and Friday July 23rd this last week (only doing the dates that you didn't already have data for, of course), and then done the "now" check today for the full range of dates. You've already said you did the last part today, so that's confirmed. The second part is more specific than your "the day that I revised my chart this week", but seems entirely consistent with you having posted the chart at 07:16 on Friday July 23rd. Does the first part seem reasonably consistent with your recollection of when you previously updated your data for the chart? (I'm obviously not expecting you to remember exact dates 5 months ago! - but does about 5 months ago seem right?)



Gengulphus


I'll deal with the date question initially as at present I can hear the sorts of sounds from Mrs Arb which suggest I am on borrowed time!

Looking at my list of prices, I see I stuck a price in the original chart of 3839p on Feb 24th. This suggests I was reacting to post regarding whether ULVR was in the buying zone or not, near that date. The Monday was 22nd Feb and sticking in a price on the 24th probably means I updated then or updated earlier and put in an "ad hoc" price on the day.
Imgur does not help as it says I uploaded the image "5 months ago" but without a date.

Anyhow, I thought you'd like to know that my dates fit in with your hypothesis.

I also notice, checking Yahoo, that they always give a Monday date if one asks for weekly price. In others words, I could ask for weekly prices on another day in the week as the prices returned are always mondays - even if that is a non-trading day.

Must go now, but I will have another bite at the cherry tomorrow, particularly with regards to your comments on how to circumvent this problem.

Arb.

Gengulphus
Lemon Quarter
Posts: 4255
Joined: November 4th, 2016, 1:17 am
Been thanked: 2628 times

Re: Unilever - Half-year Report

#430395

Postby Gengulphus » July 26th, 2021, 8:35 am

Arborbridge wrote:Looking at my list of prices, I see I stuck a price in the original chart of 3839p on Feb 24th. This suggests I was reacting to post regarding whether ULVR was in the buying zone or not, near that date. The Monday was 22nd Feb and sticking in a price on the 24th probably means I updated then or updated earlier and put in an "ad hoc" price on the day.
Imgur does not help as it says I uploaded the image "5 months ago" but without a date.

Anyhow, I thought you'd like to know that my dates fit in with your hypothesis.

Thanks - and now that you've mentioned the possibility I'd overlooked that you'd checked the price for a post, I've had a quick look and found viewtopic.php?p=389675#p389675.

Edit:

Arborbridge wrote:I also notice, checking Yahoo, that they always give a Monday date if one asks for weekly price. In others words, I could ask for weekly prices on another day in the week as the prices returned are always mondays - even if that is a non-trading day.

I don't usually use Yahoo, but checking it just now, I see that the weekly data consists of the date of the Monday at the start of the week, the opening price on the first trading day of the week (normally Monday, but bank holidays make it later some weeks), the highest price on any of the trading days of the week, the lowest price on any of the trading days of the week, and the closing price on the last trading day of the week (normally Friday, but bank holidays occasionally make it earlier). The last three prices are subject to change until the end of the week.

Gengulphus

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430415

Postby Arborbridge » July 26th, 2021, 9:47 am

Gengulphus wrote:I don't usually use Yahoo, but checking it just now, I see that the weekly data consists of the date of the Monday at the start of the week, the opening price on the first trading day of the week (normally Monday, but bank holidays make it later some weeks), the highest price on any of the trading days of the week, the lowest price on any of the trading days of the week, and the closing price on the last trading day of the week (normally Friday, but bank holidays occasionally make it earlier). The last three prices are subject to change until the end of the week.

Gengulphus


the closing price on the last trading day of the week (normally Friday, but bank holidays occasionally make it earlier).

Are you saying that the price shown in Monday is (usually) the closing price for the previous Friday?
Interesting.

It looks like your suggestion of updating my prices at the weekend up until the previous Monday could be the habit to adopt. I could add the lowest price for the week by visual inspection and put this in manually, but that increases my labour and I'm not sure it is worth it. After all, this exercise was only started as a rough and ready indication of whether ULVR (or other companies)
was in the ball park for "as good as it gets". Adding my buy points and comparing dividend increases with RPI where added as further points of interest.

None of this is mission critical, but it's nice to get it right.

Arb.

Gengulphus
Lemon Quarter
Posts: 4255
Joined: November 4th, 2016, 1:17 am
Been thanked: 2628 times

Re: Unilever - Half-year Report

#430468

Postby Gengulphus » July 26th, 2021, 12:29 pm

Arborbridge wrote:
Gengulphus wrote:I don't usually use Yahoo, but checking it just now, I see that the weekly data consists of the date of the Monday at the start of the week, the opening price on the first trading day of the week (normally Monday, but bank holidays make it later some weeks), the highest price on any of the trading days of the week, the lowest price on any of the trading days of the week, and the closing price on the last trading day of the week (normally Friday, but bank holidays occasionally make it earlier). The last three prices are subject to change until the end of the week.

Gengulphus


the closing price on the last trading day of the week (normally Friday, but bank holidays occasionally make it earlier).
Are you saying that the price shown in Monday is (usually) the closing price for the previous Friday?

No, for the following Friday.

E.g. if you ask for weekly data, you need to read the current line for last week, which says:

Date          Open       High       Low        Close*     Adj. close**   Volume
19 Jul 2021 4,356.50 4,388.00 4,019.68 4,144.00 4,144.00 24,069,710

as if it said:

Dates of week    Open       High       Low        Close*     Adj. close**   Volume
19-23 Jul 2021 4,356.50 4,388.00 4,019.68 4,144.00 4,144.00 24,069,710

But if you were to look at them on this coming Thursday, for instance, the line starting with the date 26 Jul 2021 would be for the dates of a partially complete week: 26-28 Jul 2021. And if you try on a Monday before the market has closed (as I've just tried), what you get is:

Date          Open       High       Low        Close*     Adj. close**   Volume
26 Jul 2021 - - - - - -

Presumably it will be updated for today's open, high, low and close prices sometime after the market closes today, then after the market closes tomorrow, it will be updated by:

* Leaving the opening price unchanged (today will still be the start of the week so far).
* Replacing the high price by tomorrow's high price if it is higher than today's, but leaving it unchanged otherwise (since the high for the week so far will be the highest price seen either today or tomorrow).
* Replacing the low price by tomorrow's low price if it is lower than today's, but leaving it unchanged otherwise (since the low for the week so far will be the lowest price seen either today or tomorrow).
* Replacing the closing price by tomorrow's closing price regardless (since tomorrow will be the new end of the week so far).

And then similar updates will be made after the market close on each of Wednesday, Thursday and Friday, after which the data for this week (26-30 Jul 2021) will be complete and it will move on to assembling the data for next week.

Gengulphus

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430634

Postby Arborbridge » July 27th, 2021, 7:39 am

Gengulphus wrote:Are you saying that the price shown in Monday is (usually) the closing price for the previous Friday?
No, for the following Friday.

E.g. if you ask for weekly data, you need to read the current line for last week, which says:

Date          Open       High       Low        Close*     Adj. close**   Volume
19 Jul 2021 4,356.50 4,388.00 4,019.68 4,144.00 4,144.00 24,069,710

as if it said:

Dates of week    Open       High       Low        Close*     Adj. close**   Volume
19-23 Jul 2021 4,356.50 4,388.00 4,019.68 4,144.00 4,144.00 24,069,710


Gengulphus


Gosh, how extraordinary. I think I get it now, but it makes all the dates on the chart incorrect by five days - not that that would show up on the small scale.
i.e. when the table tells me 19th July was 4144p it should be read as "the closing price on the friday of the week commencing 19th July was 4144p". Whereas the closing price on 19th July was actually 4346p, I notice from the daily data.

Updating at the weekend gives me the Friday closing price (assuming they've updated the website) but I'd paste and copy the Monday date. Hmm - well, I'll live with that.

Arb.

Arborbridge
The full Lemon
Posts: 10371
Joined: November 4th, 2016, 9:33 am
Has thanked: 3601 times
Been thanked: 5227 times

Re: Unilever - Half-year Report

#430656

Postby Arborbridge » July 27th, 2021, 9:48 am

Gengulphus wrote:And then similar updates will be made after the market close on each of Wednesday, Thursday and Friday, after which the data for this week (26-30 Jul 2021) will be complete and it will move on to assembling the data for next week.

Gengulphus


Going back to the original "problem" I've altered the run of numbers for March 20, the "anomalous" peak, but inserting the low of 3583p which occurred on March 16th and also the end of week price of 4138p on the 20th.
This now makes the first peak higher than the second peak, but it's a curiosity as the peak was so sharp as to be impractical to buy. Might be more "normal" to replace it with the closing price of 3726p to be consistent with all the other charts.

Anyhow, all I really need to know is that my buy of May 20 missed the peak!

And if one can buy ULVR at 3.75-4% (historic) that's about as good as it usually gets - which was the point of the chart exercise.

Arb.

ADrunkenMarcus
Lemon Quarter
Posts: 1584
Joined: November 5th, 2016, 11:16 am
Has thanked: 672 times
Been thanked: 479 times

Re: Unilever - Half-year Report

#430731

Postby ADrunkenMarcus » July 27th, 2021, 3:00 pm

Arborbridge wrote:And if one can buy ULVR at 3.75-4% (historic) that's about as good as it usually gets - which was the point of the chart exercise.


I believe I was around the low end of that range when I initiated my holding, building it up in two tranches in 2013 - certainly the second tranche. Even at today's levels it has shown a healthy return. I view the ULVR and DGE's as valuable stalwarts for an HYP, if you can get them at a decent entry point.

Best wishes


Mark.


Return to “HYP Practical (See Group Guidelines)”

Who is online

Users browsing this forum: No registered users and 12 guests