LIC HFL Assistant: How to Crack Missing Data in DI

The LIC HFL Assistant recruitment run by the Life Insurance Corporation of India has turned into one of the most competitive entry-level insurance exams in the country, and a growing share of candidates now sit for it from Australian time zones. The prelims paper keeps quantitative aptitude and data interpretation in the same block, so you rarely get to settle into a comfortable rhythm. You will move from a percentage-based table straight into a cash-flow statement, and then into a set where the table itself arrives with holes in it.

Those holes are the subject of this piece. A missing data question inverts the usual arrangement of a DI set. Instead of handing you a complete table and asking you to read a value off it, the examiner asks you to work backwards and supply the value that makes the whole table consistent. The reassuring part is that the answer is never a guess. It is always anchored somewhere in the stimulus: a total row, a column total, a percentage relationship stated in the stem, or a figure you calculated yourself two questions earlier.

These items punish the same weakness twice: speed without structure. Candidates who read tables top-to-bottom, left-to-right, hunting for the blank, burn valuable minutes on arithmetic that a total row would have handed over in seconds. Candidates who are quick but careless produce a value that does not belong to the table at all — three times larger than its neighbours, or a percentage above 100 — and the options quietly confirm the error.

What follows is a repeatable routine, three reconstruction methods, a percentage-chain method that works in both directions, and a short list of traps specific to this paper. Where the examples use Australian data, that is deliberate. Branch networks of the big four banks, the Reserve Bank's cash rate, the 10% GST and dollar magnitudes will sit more comfortably than notional rows about imaginary companies. If you are revising in Sydney, Melbourne or Brisbane, keep your practice tables in the units you actually read: thousands and millions, not lakh and crore.

What the missing data item is really asking

You will meet three broad versions in the LIC HFL Assistant prelims and mains. The first is a blank cell inside a table that carries row totals, column totals, or both. The second is a missing total, where every cell is present but the bottom line has been removed. The third is a relationship-based item, where the stem tells you that one category is a fixed proportion or multiple of another and asks for a figure that is nowhere in the table.

A fourth variant appears in the statement-based portion, where you are asked whether a claim can be verified from the data, and the honest answer is that it cannot. That last type is the cruellest, because it rewards a candidate who has actually mapped the information available.

Whatever the disguise, the question resolves into one of three anchors. A total anchor means some sum is given and your job is subtraction. A ratio anchor means the stem supplies a proportional relationship, so the missing figure is a scaling exercise. A statement anchor means the stem itself carries the only relevant number, and everything else is decoration. Recognising which anchor you are holding is worth more than any calculator technique, because it tells you in advance whether the answer will be a tidy figure or an awkward decimal.

Mind the set design as well. DI sets are frequently linked, where the value you compute in the fourth question is required in the sixth. A missing data item in the middle of a set can therefore depend on an earlier answer, so skipping ahead and guessing blind throws away marks that were effectively handed to you. Negative marking in the prelims sits at a quarter mark per incorrect response, which makes an informed attempt always worth it, and a half-knowledge reconstruction with two unknowns and one equation is not informed enough.

A fixed routine before you touch the arithmetic

Before any calculation, spend sixty to ninety seconds orienting the table. Establish the units: individual counts, thousands, or millions. Establish the currency and the period, and check whether the fiscal year runs July to June, as it does in Australia, or April to March, as it does in India, because a table that mixes the two is a favourite examiner construction. Establish the denominator of any percentage column, whether it is the row total, the column total, or the grand total. Most DI errors in this paper are not arithmetic errors at all; they are errors of framing.

Then catalogue the anchors on your rough sheet in shorthand: T for a total, R for a row, C for a column, and a short equation for anything the stem states, such as A = 1.2B or X = 30% of total. Only then locate the blank and count how many unknowns you face against how many equations you hold. One equation and one unknown means you have a solvable problem; two unknowns and one equation means you have a "cannot be determined" answer, and no amount of effort will change that.

If your instinct is to guess whenever the information feels thin, work on telling solvable and under-determined questions apart. The data sufficiency accuracy drills in our PGCET MBA guide train exactly that reflex, and they transfer cleanly to this paper.

Finally, decide the method before you compute: subtraction from a total, ratio scaling, the difference method for a stated percentage change, or a percentage chain. Estimate the magnitude of the answer first, then check the reconstructed figure against its neighbours. A table of Australian bank branch numbers should not produce a cell of 300 when the rest of the row sits between 20 and 60.

Rebuilding missing rows and columns

Consider a table of retail home loan approvals, in thousands, for Commonwealth Bank, ANZ, NAB and Westpac across Sydney, Melbourne and Brisbane, with each row and column totalled. Sydney shows 44, 32, a blank and 18 against a total of 120. The reconstruction is immediate: 120 − (44 + 32 + 18) = 26.

Now suppose instead that the Melbourne cell in the NAB column is blank, while the NAB column total reads 56. With Sydney at 26 and Brisbane at 8, the Melbourne figure is 56 − 26 − 8 = 22. The same table then supplies a missing row total for Brisbane, which is 28 + 16 + 8 + 12 = 64, and can be verified from the grand total: 280 − 120 − 96 = 64. Two independent paths, one answer.

The fourth case is the hardest. When the table carries no total for the row or column containing the blank, you must reach for a ratio stated in the stem or in a footnote, such as one bank's Sydney approvals running at 1.5 times another's Melbourne approvals, and scale accordingly. Resist the temptation to assume an even distribution across cities unless the data itself supports it; proportional reasoning works only when a proportional relationship is actually given. If none exists, the correct response is to recognise that the value cannot be determined, which is frequently the intended answer in mains-level sets.

Once you have a candidate value, interrogate it. Does the row's internal shares still look plausible, with Sydney roughly 42 per cent of CBA's approvals, as 44 out of 104 suggests? Does the reconstructed figure sit within the range of the two cells flanking it? Do the options confirm you, with values like 18, 22, 26 and 30 leaving only one that reconciles the table? This two-minute audit catches more errors than any amount of extra practice.

Percentage chains and missing bases

Percentage relationships are where missing data questions become genuinely elegant, because they often work in both directions. If FY24 is 15 per cent higher than FY23, you can move forward by multiplying by 1.15 and backward by dividing by 1.15. The trap is treating a backward step as subtraction: dividing by 1.15 is not the same as removing 15 per cent from the given figure, and the difference grows with the size of the number.

Take a headcount table across three financial years. FY22 stands at 20,000, FY23 is 10 per cent higher at 22,000, and FY24 is 15 per cent higher again at 25,300. Blank out FY23 and you recover it as 25,300 ÷ 1.15 = 22,000. Blank out FY22 and you divide by the compounded factor, 25,300 ÷ (1.10 × 1.15) = 25,300 ÷ 1.265 = 20,000. Notice what this exposes: FY24 sits 26.5 per cent above FY22, not 25 per cent, because the two increases land on different bases. Examiners build items around exactly that error.

Shared-base items deserve the same discipline. If ANZ represented 30 per cent of a bank's retail headcount and the headcount was given, the total is that figure divided by 0.30. A second category at 45 per cent of the same total is then 0.45 × (given ÷ 0.30), or 1.5 times the first category. The same arithmetic appears whenever a table mixes GST-inclusive and GST-exclusive values, and here Australian candidates have a genuine edge: the Australian GST is 10 per cent, so the inclusive factor is 1.10, while Indian GST at 18 per cent gives 1.18. Solve in the currency and the tax regime printed in the table, not the one you remember.

One further distinction to keep sharp is percentage points against percentages. The Reserve Bank lifted the cash rate target from 0.10 per cent in mid-2022 to above 4 per cent within a year, a move of more than four percentage points that sounds modest but changes the arithmetic on a $600,000 variable loan from roughly $50 a month in interest to roughly $2,175. On a DI table, "fell by 2 per cent" and "fell by 2 percentage points" lead to different answers, and the stem tells you which one is in play.

Traps, timing and the marks you can still win

Units cause more lost marks than any other factor. Lakhs and crores are unfamiliar currency blocks to many people studying in Australia, and thousands, millions and billions are not, so convert your practice tables early. A figure of 1.2 million in a table stated in thousands means 1,200, and a candidate who skips that step will pick the wrong option with full confidence. The same applies to a table headed in AUD while the stem quotes dollars in another currency: convert only if the question asks for it, and remember the exchange rate moves every day, which makes it a poor anchor for anything.

Two answer types get confused constantly. "Cannot be calculated" means you lack the arithmetic route but the value exists in the data. "Cannot be determined" means the value does not exist uniquely. A statement such as "A's approvals exceeded B's" can often be verified from a ratio without knowing either absolute figure, so treat the "cannot be calculated" option with suspicion when the relationship is explicit.

On timing, budget roughly ninety seconds per set, and treat a missing data item that resists the routine for more than that as a skip. A quarter mark of negative marking means a wild guess is a bad trade, while a structured attempt has positive expected value. Work through the daily assignments, mock tests and current affairs material on the Competitive Success site, and if you are outside India, remember that Indian Standard Time runs four and a half hours behind Sydney in standard time, so plan your evening sessions and doubt sessions accordingly. Candidates who can attend in person can use the library and classroom support at the Sheshadripuram centre in Bangalore.

Finally, keep an error log that names the failure rather than the topic: anchor missed, unit error, compounding error, arithmetic slip. An error log turns a long practice cycle into a short repair list, which is how the marks actually come back. If your plan also covers Karnataka's state-level recruitment, the KPSC FDA Kannada guide handles the language paper in the same rhythm, and the two preparations complement each other well. Read tables the way a batter reads a scorecard at the MCG: the answer is usually sitting in the frame already, and the skill is spotting which part of the frame to look at.